NeurIPS 2026 Schedule

Filter Events

THU 10 DEC

1 a.m.

Atlanta Invited Talk:

Physical AI: Technology, Trust, and How Not to Lose Your Cool

Ayanna Howard

(ends 2:00 AM)

2 a.m.

Break:

Coffee & Snacks

(ends 2:30 AM)

Atlanta Poster Session 1 [2:00-5:00]

Posters 2:00-5:00

$\epsilon$-Good Action Identification in Fixed-Budget Monte Carlo Tree Search

$f$-GRPO & Beyond: Divergence-Based Reinforcement Learning Algorithms for General LLM Alignment

3DCodeBench: Benchmarking Agentic Procedural 3D Modeling Via Code

3DSPA: A 3D Semantic Point Autoencoder for Evaluating Video Realism

A$^2$IQL: Adaptive Asymmetric Implicit Q Learning for Automated Warehouse Consolidation

Accelerated last-iterate convergence of Extragradient via power-law stepsizes

A Communication-Theoretic Framework for LLM Agents: Cost-Aware Adaptive Reliability

ACQueReLlo: Alignment-Aware Constrained Quantization via Reinforcement Learning for Large Language Models

Adaptive Inference for Functionals of M-Estimands

AdaST: Adaptive Coupling for Spatial-Temporal Forecasting

A general kernel framework for non-CND distance measures using $|\mathcal{D}|$-dimensional sparse landmark embeddings

AgentKVShift: Efficient KV Cache Reuse for Agentic Memory Systems

A Large-Scale Multi-Source Dataset Linking Hacker Community Discourse to the CVE Vulnerability Lifecycle

Align and Distill: Unifying and Improving Domain Adaptive Object Detection

Aligning Language Models with Selective Prediction

Alleviating Hallucination with Training-Free Uncertainty-Guided Steering

A Method-Class Divide in Sub-4-Bit Quantization:\\Iterative vs.\ Single-Pass Sensitivity to Data Composition

Amortized Vine Copulas for High-Dimensional Density and Information Estimation

Amplitude Decoupling in Gaussian Process Training: Exact Decomposition, Pole Cancellation, and Evaluation-Efficient Optimization

Anchoring LLM-based Chest X-ray Report Generation via Diffusion Language Planning

Approximation Guarantees for Robust Aggregation in Federated Learning

A Refined Sample-Complexity Analysis of Robust Policy Optimization under Decaying Actor Stepsizes

A Set-Sequence Model for Time Series

ASSET: Acquisition-Sensitive Subspace Estimation for Test-Time Adaptation of Medical VLMs

A Statistical Theory of Gated Attention through the Lens of Hierarchical Mixture of Experts

A Tight Hierarchy For Chain of Thought

AtlasVid: Efficient Ultra-High-Resolution Long Video Generation via Decoupled Global-Local Modeling

A Transformer-Derived Iterative Preconditioner

AUTOMEM: Automated Learning of Memory as a Cognitive Skill

BACE: Behavior-Adaptive Connectivity Estimation from Multi-Region Neural Recordings

Bayes-Sufficient Compression Is Not Enough: How Communication Helps in Multi-Agent Systems?

B-CALM: Bias-Limited Bayesian Borrowing for RCT-Anchored Treatment Effects under Covariate Mismatch

Benchmarking Multi-Modal Graph-based Social Media Popularity Prediction

Benchmarking Multimodal Mathematical Reasoning with Explicit Visual Dependency

Beyond Domains: Reusing Web Skills via Transferable Interaction Patterns

Beyond Linear Activation Steering: Invertible Latent Transformations for Controlling LLM Behavior

Beyond LLM-Based Reasoning: Lightweight GNNs for Agent Failure Attribution

Beyond Local Neighborhoods: Fractional Diffusion with Levy Flights on Simplicial Complexes for Link Prediction

Beyond Maximum Likelihood: Variational Inequality Estimation for Generalized Linear Models

Beyond Score: A Dataset for Joint Action and Score Predictions in 2v2 Sports

Beyond Truthfulness: Evaluating Honesty in Large Language Models

Beyond Unit-Circle Eigenvalues: Invariant Bases for Stable State Space Dynamics

BigCell: Generating Gigapixel Whole-Slide Images

BioBlobs: Unsupervised Discovery of Functional Substructures for Protein Function Prediction

Bounding Global and Local Compression Error of Signal Parameterizations

Breakeven complexity: A new perspective on neural partial differential equation solvers

Bridge Graphical Models: Coupling, Projection, and Current-Preserving Dynamics for Generative Modeling

C3VD-DEFCOL: A Deformable Colonoscopy Dataset with Time-Resolved 3D Ground Truth and Realistic Appearance

Calibeating Prediction-Powered Inference

Causal Concept Explanations for Deep Neural Models

Causal Survival Forests with Negative Controls

CC-GS: Low-Memory 3D Gaussian Splatting Training via CPU-GPU Block-Wise Context Compositing

CD-RCM: Generalizable Continuous-Depth Novel View Synthesis for Reflectance Confocal Microscopy

Certified but Private: Scalable Zero-Knowledge Proofs for the Formal Verification of Neural Networks

C-GRPO: Conformal Group Relative Policy Optimization

Chain-of-Route: State-Aware LLM Routing for Multi-Turn Conversations

ClusQuant: Mitigating Outliers with Clustering-Based Representations for Low-Precision LRMs

CM2: Reinforcement Learning with Checklist Rewards for Multi-Turn and Multi-Step Agentic Tool Use

Cognitive constraints and thalamocortical architecture explain systematic biases and neural signatures in human hierarchical decision-making

COMPLLLM: Fine-tuning LLMs to Discover Complementary Signals for Decision-making

Concorde: Geometry-Aware Link Prediction via Decoupled Energy Minimization

CONSTRAINER: Promptable Graph-Structured Optimization via Constraint Conditioning

Contrastive Pretraining Scales Agentic Exploration

Contrastive Representation Shaping for LLM Unlearning

Counterfactual Debugging the World Model Transfer Gap

Counterfactual Rollout Replay: Forkable Environments as Free Process Rewards for Software Engineering Agents

Coupling Models for One-Step Discrete Generation

Crafting Reversible SFT Behaviors in Large Language Models

Cross-Cell-Line Perturbation Prediction Needs Controls

Curvature Beyond Positivity: Greedy Guarantees for Arbitrary Submodular Functions

CytoWave: Perturbation-Centric Pretraining for Single-Cell Response Prediction

DarkVGGT: Seeing Through Darkness Using Thermal Geometry without Daylight Tax

Data Auctions for Retrieval Augmented Generation

Data-Constrained Language Model Pretraining: Improved Regularization and Scaling Laws

DD-Ranking: Rethinking the Evaluation of Dataset Distillation

Debiased DPO for Diffusion Models

Differentiable Range-Partition Entropy for Entropy-Sensitive Geometric Algorithms

DiPMInd: Distance profile based mutual independence testing for random objects

Dominant-Layer ZO: A Single Layer Dominates Zeroth-Order Fine-Tuning of LLMs

Do More Modalities Always Help? A Geometric Perspective on Missing-Modality Robustness

Do Reasoning LLMs Refuse What They Infer in Long Contexts?

Drift Flow Matching

DualKV: Shared-Prompt Flash Attention for Efficient RL Training with Large Rollouts and Long Contexts

DynaTokens: Teaching Dynamics to Camera-Controlled Video Models at Test Time

Early Signals, Strong Decisions: Prefix-Guided Sampling for Parallel Test-Time Scaling

Economy of Minds: Emerging Multi-Agent Intelligence with Economic Interactions

Efficient Dynamic Algorithms for Graph Neural Networks with Non-Linearity

EgoSurgHands: An Egocentric 3D Hand Pose Dataset & Benchmark for Open-Surgery Training

ELPAC: Endpoint-Anchored Latent Progression with Stage-Varying Multimodal Coordination

Enabling Preference-driven Unlearning in Few-step Distilled Text-to-Image Diffusion Models

Ensembits: an alphabet of protein conformational ensembles

EntityBench: Towards Entity-Consistent Long-Range Multi-Shot Video Generation

Evaluating Deployable Inference-Time Error Prediction in Vision MoEs

Evaluating Whether LLMs Can Reliably Connect the DOTs?

Everything at Every Scale: Scale-Invariant Diffusion with Continuous Super-Resolution

Explanations over Graphs: An Agent Architecture for IT Enterprise Diagnostic Tasks

ExploitGym: Can AI Agents Turn Security Vulnerabilities into Real Attacks?

Exploring Advertising Manipulation in Diffusion Image Generation

Express Language Modeling

FASD: Hardware Acceleration for Multi-AI-Agent Discussion

Faster Rates For Federated Variational Inequalities

FedRSPO+: A Heterogeneity-aware Algorithm for Decision-focused Federated Learning

Finding Interpretable Prompt-Specific Circuits in Language Models

Fine-Tuning Improves Information Conveyance in Language Models

Finite-Sample and Communication-Efficient Networked Information Aggregation

First-Order Regret for Online Convex Optimization with Memory and Online Nonstochastic Control

FLAME: Adaptive Mixture-of-Experts for Continual Multimodal Multi-Task Learning

FLARE: Diffusion for Hybrid Language Model

Flow Annealing Posterior Sampling for Function-Space Regression and Inverse Problems

FLUX: Geometry-Aware Longitudinal Flow Matching with Mixture of Experts

From History to State: Constant-Context Skill Learning for LLM Agents

From Likelihood Convergence to Parameter Convergence in POMDPs

From POMDP Theory to Deep RL with Particle Filters

Fully Distributed Tâtonnement for Chores Markets

Gaeta-Lie Neural SDEs: Symmetry-Regularized Learning of Stochastic Dynamics

Gated DeltaNet-2: Decoupling Erase and Write in Linear Attention

GEAR: A GPU-Accelerated Global Solver for Nonlinear Programs via Linear Bound Propagation

Generalization Measures for Deep Learning Should Be Audited for Fragility

Generative Modeling via Drifting

GeoTransolver: Learning Physics on Irregular Domains using Multi-scale Geometry Aware Physics Attention Transformer

gHAWK: Structural Encoding for Scalable Training of Graph Neural Networks on Knowledge Graphs

Global Convergence of Four-Layer Matrix Factorization under Random Initialization

Gradient Descent on Two ReLU Neurons: Global Landscape and Bifurcation Dynamics

Harness the Memory: A Holistic Evaluation of Memory Substrates in Memory Agents

How Accurately Can a Gaussian Approximate Stochastic Approximation Iterates?

How New Strategies Emerge in RL Post-Training: A Controlled Study

HPC-Bench: A Comprehensive Benchmark for High Performance Computing Codes

Hybrid-LoRA: Bridging Full Fine-Tuning and Low-Rank Adaptation for Post-Training

HYVE: Hybrid Views for LLM Context Engineering over Machine Data

Improved techniques for fine-tuning flow models via adjoint matching: a deterministic control pipeline

IncAgg: Efficient Memory-Enhanced Graph Learning via Incremental Aggregation

Inference for Many Quantiles under Local Differential Privacy

Inference-Time Search Using Side Information for Diffusion-Based Image Reconstruction

Information-Geometric Forward Policy Training in GFlowNets

Information Templates: A New Paradigm for Intelligent Active Feature Acquisition

Instance-Adaptive Online Multicalibration

Inverting Retargeting: Humanoid Datasets Remember Their Operators

Iterative Critique-and-Routing Controller for Multi-Agent Systems with Heterogeneous LLMs

KerONet: A Single Softmax Readout Suffices for Physics-Informed Operator Learning

KV-PRM: Efficient Process Reward Modeling via KV-Cache Transfer for Multi-Agent Test-Time Scaling

L2-Bench: An Evaluation Benchmark for Measuring LLM Capabilities in Second Language Education

Label-Free Consistency Correction for Weak-to-Strong Generalization

LARK: Learnability-Grounded Trajectory Selection for Efficient Reasoning Distillation

LASER: Latent Space Adjoint Matching for Support Constrained Entropy Regularized Offline RL

Latent Generative Solvers for Generalizable Long-Term Physics Simulation

LatentRouter: Can We Choose the Right Multimodal Model Before Seeing Its Answer?

Lattice Deduction Transformers

Learnable Chernoff Baselines for Provable Inference-Time Alignment

Learning Actionable Information Landscapes for Multimodal Active Sensing in Hawkmoths

Learning Digital Twins under Drift: Optimal Tracking Rates for Non-Stationary Dynamical Systems

Learning Distributions from Multiple Data Providers

Learning Fractional-Order Dynamics from a Single Trajectory

Learning Orthogonal Multi-Index Models Beyond Small Initialization: Incremental Learning, Competitive Dynamics and Symmetry

Learning Provable Neural Network Observer for Uncertain Dynamical Systems

Learning to Correct Geometry in Generated Videos

Learning to Cut: Reinforcement Learning for Benders Decomposition

Learning What Evaluators Value: A Reliable Approach to Modeling Evaluator Preferences

LESSViT: Robust Hyperspectral Representation Learning under Spectral Configuration Shift

Leveraging Soft Prompts for Privacy Attacks in Federated Prompt Tuning

LiFT: Lifted Inter-slice Feature Trajectories for 3D Image Generation from 2D Generators

LIPAR: Latent Inter-Frame Pruning with Attention Recovery

LLM Judge Validation Under Sparse Overlap: From Inference to Design

Long-Context Language Models Require Extreme Sparsity in Context Dimension

LongMINT: Evaluating Memory under Multi-Target Interference in Long-Horizon Agent Systems

LowRankArena: A Standardized Evaluation Platform for SVD-Based LLM Compression

Margin Dynamics for Large Language Model Alignment

MaxIM: Maximally Informative Incremental Summarization via Reinforcement Learning

Mini Amusement Parks (MAPs): A Testbed for Modelling Business Decisions

Mission Impossible: Diagnosing and Fixing Non-Operative Instruction Following in Image Editing

Mitigating Data Heterogeneity Effect in Client-Reshuffling-Based Federated Learning

MITO: A Millimeter-Wave Dataset and Simulator for Non-Line-of-Sight Perception

MonoPhysics: Estimating Geometry, Appearance, and Physical Parameters from Monocular Videos

Move on Muon : A Hamiltonian probability gradient flow perspective of Muon optimizer

MSConsensus: A Hundred-Million-Scale, Batch-Effect–Suppressed Dataset and Benchmark for Proteomics Machine Learning

Multi-site PPG: An In-the-Wild Physiological Dataset from Emerging Multi-Site Wearables

NeuralFieldManifold: Reconstruction of LFP manifold with Lag Embedding

nnTrace: Detecting and Localizing Silent Bugs in Distributed Training

No One Knows the State-of-the-Art in Geospatial Foundation Models

Not All Spines Are Created Equal: How CT Segmenters Fail on Lumbosacral Transitional Vertebrae

On Computing Diverse Solutions in the Earth Movers Distance

One for All: A Non-Linear Transformer can enable Cross-Domain Generalization for In-Context Reinforcement Learning

On the Divergence of Differential Temporal Difference Learning without Local Clocks

On the Optimal Sample Complexity of Offline Multi-Armed Bandits with KL Regularization

On the Provable Emergence of Hierarchical Concept Structure in CLIP Embeddings

Optimal In-Context Learning of Autoregressive Processes under Heterogeneous Second-Order Moments of the Prompts

Order-Optimal Sample Complexity for Distribution Learning via Flow Matching

Ordinal Geometry Complements Reconstruction: Diagnosing Planning with Compressed Value Functions

Orthrus: Memory-Efficient Parallel Token Generation via Dual-View Diffusion

PAC Learning with Bandit Feedback: Sharp Sample Complexity in the Realizable Setting

PAIR-CI: Calibrated Conditional Independence Testing for Causal Discovery with Incomplete Data

Palette: A Modular, Controllable, and Efficient Framework for On-demand Authorized Safety Alignment Relaxation in LLMs

Penalty-Based First-Order Methods for Bilevel Optimization with Minimax and Constrained Lower-Level Problems

Perceive-then-Plan: Layout-as-Policy for Monocular 3D Scene Layout Estimation

Physics Unrolled Neural Operator for Wireless Field Modeling

PithTrain: A Compact and Agent-Native MoE Training System

PixelDiT2: Representation-Grounded Pixel Diffusion Transformers

PLATO: Pointer Learner for Agent and Task Openness

Policy Regret Minimization in Partially Observable Markov Games

PORTool: Importance-Aware Policy Optimization with Rewarded Tree for Multi-Tool-Integrated Reasoning

PosteriorBench: From Point Estimates to Posterior Matching in Evaluating Generative Inverse Solvers

Preference Conditioned Multi-Objective Reinforcement Learning: Decomposed, Diversity-Driven Policy Optimization

Preferential dynamic modeling with forward-backward smoothing

Primal-Dual Flow Matching for Sample-Wise Constrained Generation

Privacy-Preserving Retrieval-Augmented Generation with Plausible Deniability

Provably Efficient Representation Learning for Low-Rank CMDPs

Quantifying Centrality for Complex Data

Quantum Safe Stochastic Linear Bandits

Quest: Training Frontier Deep Research Agents with Fully Synthetic Tasks

Rethink Action Chunking in VLA Through Human Motor Control

Rethinking Long-Video Efficiency: A Joint Allocation Perspective on Frames, Pixels, and Front-End Latency

Rethinking Psychometric Evaluation of LLMs: When and Why Self-Reports Predict Behavior

Riemannian Bilevel Optimization under the Polyak–Łojasiewicz Condition

Robust and Hard-to-Remove GNN Watermarking via Topological Invariant Perception

Robust Hopfield Decision Transformer

s2n-bignum-bench: A practical benchmark for evaluating low-level code reasoning of LLMs

Safeguarding LLMs via Model-Agnostic Latent Safety Signals from Dark Knowledge

Safe Score Matching: Diffusion Policies with Hamilton-Jacobi Reachability for Online Safe Reinforcement Learning

Same Signal, Opposite Meaning: Direction-Informed Adaptive Learning for LLM Agents

SAPO: Step-Level Skill-Augmented Policy Optimization for Multi-Turn LLM Agents

Scalable Neural Safety Certification via Monotonicity

Scalable Variational Bayesian Fine-Tuning of LLMs via Orthogonalized Low-Rank Adapters

Scaling Reward Modeling without Human Supervision

SCOT: Multi-Source Cross-City Transfer with Optimal-Transport Soft-Correspondence Objectives

Search-Augmented Masked Diffusion Models for Constrained Generation

SeeSE3: The Emergence of 3D Space in Vision Features

Self-Compacting Language Model Agents

Self-Trained Verification for Training- and Test-Time Self-Improvement

SGD at the Edge of Stability: The Stochastic Sharpness Gap

Signed Rectified Flow: Negativity Controlled Generation

Simulation-Informed Diffusion for Decentralized Multi-robot Motion Planning

SlackBench: Benchmarking Agents on Collaborative Projects Grounded in Real Code Repositories

Soteria: Formally Verified Planning with Runtime Enforcement for Safe LLM Agents

Specialist Mediators: Causal Localization of Dense Fine-Tuning

Specialists Hold, Generalists Discount: Asymmetric Equilibrium in LLM Routing Auctions

Spike-to-Field Mechanisms of Turbulence-Like Dynamics in Spatial Spiking Neural Networks

SPIRIT: Speed-Driven Online Adaptation for Self-Speculative Decoding

Split and Bridge: Multimodal Generation via Diffusion Bridging

Stability-Constrained Regime-Aware Forecasting for Heterogeneous Panel Time Series

Stable GFlowNets with Probabilistic Guarantees

Statistical Matching via Schr\"odinger Bridge beyond Conditional Independence

Stealthy World Model Manipulation via Data Poisoning

Stein Kernelized Molecular Dynamics for Active Learning of Interatomic Potentials

Stop Calling It Reinforcement Learning in Language Models Without Clear Improvement Claims: Decision-Process Cards as a Reporting Standard

Structural Rationale Distillation via Reasoning Space Compression

Structured Masked Diffusion for Joint Multiuser Decoding

Sublinear Time Quantum Sensitivity Sampling

SWE-Protégé: Learning to Selectively Collaborate With an Expert Unlocks Small Language Models as Software Engineering Agents

Synthetic Web: Benchmarking Language Agents under Adversarial Search Ranking

TailGuard: Subgroup Tail Coverage Theory for Safe LLM Alignment

Talk Less, Work More: Communication-Efficient Decentralized Stochastic Approximation

Teaching LLMs to Recommend and Defer in Underrepresented Epilepsy Care

Test-Time Defense Against Adversarial Attacks via Stochastic Resonance of Latent Ensembles

TextRegion: Text-Aligned Region Tokens from Frozen Image-Text Models

TGPO: Temporal Grounded Policy Optimization for Signal Temporal Logic Tasks

Theory Guided and Interpretable Neural Operator Design for Partial Differential Equation Learning

TherapyGym: Evaluating and Aligning Clinical Fidelity and Safety in Therapy Chatbots

Tiny but Trusted: Efficient Vision-Language Reasoning for Time-Series Anomaly Detection

To discretize continually: Mean shift interacting particle systems for Bayesian inference

Tokenizer Choice Shapes Generalization in State-Centric Learning for Planning

Toward Cognitive Supersensing in Multimodal Large Language Models

Towards Direct Latent-Space Synthesis for Parallel Branches in LLM-Agent Workflows

Towards Fair Graph Generation Without Sensitive Attribute

TraceML: An Empirical Analysis of Human-Agent Planning in Machine Learning Development

Transfer Entropy as a Measure of Information Flow in VLMs and LLMs

Transformers Provably Learn Graph Search: Training Dynamics and the Exponential Power of Depth

TransmissiveGS: Residual-Guided Disentangled Gaussian Splatting for Transmissive Scene Reconstruction and Rendering

Tune-Up Open-Weight CLIP: Optimization Framework for Self-Supervised Fine-tuning of CLIP

Two-Sided Learning in Matching Markets with Interviews

Two Speeds of Learning: A Representation-Readout Decomposition of Grokking and Double Descent

UDT: Reconciling U-Nets and Diffusion Transformers with Data-Adaptive Token Reduction

Unified Generative-Predictive Modeling for 4D Scene Understanding

UniPro: Unified Multi-Mode Medical Image Segmentation from 2D Images to 3D Volumes via Propagation

Urgency-Aware Autoregressive VLMs for Unanticipated Healthcare Occurrences

Variational Trajectory Optimization of Anisotropic Diffusion Schedules

VIGOR: Benchmarking Visual Rationale Correctness in Multimodal Large Language Models

VIPBench: A Human-Aligned Benchmark for Voice Identity Perception in the Age of Voice Cloning

VOID: Backdoor Injection through Knowledge Vacuity in Federated Unlearning

WebNavigator: Global Web Navigation via Interaction Graph Retrieval

Weighted Conformal Clustering

What Makes a Good Path? Factoring Manifold Support and Path Geometry

What Should a Streaming Video Model Remember?

When Does Knowing the State Help? Diagnosing Process vs. Outcome Reward Design

When Do Multi-Agent Systems Help? An Information Bottleneck Perspective

When Scores Conflict with Preferences: Calibrated Drift Control for Heterogeneous DPO

When Think-with-Image Meets Safety: What Determines Multimodal Jailbreak Robustness?

Why Muon Outperforms Adam: A Curvature Perspective

Your Embedding Model Is SMARTer Than You Think

Z0-Inf: Zeroth Order Approximation for Data Influence

(ends 5:00 AM)

4:30 a.m.

Break:

Retail Lunch

(ends 6:30 AM)

5:30 a.m.

Affinity Poster Session:

Affinity Joint Poster Session

(ends 8:00 AM)

7:30 a.m.

Atlanta Orals 1A Reliable & Safe LLM Agents and Reasoning [7:30-8:30]

Orals 7:30-8:30

[7:30] AgentAbstain: Do LLM Agents Know When Not to Act?

[7:30] Fisher-R1: Training LLM Agents for Reliable Hypothesis Testing

[7:30] Soteria: Formally Verified Planning with Runtime Enforcement for Safe LLM Agents

[7:30] Escaping the Cognitive Well: Efficient Competition Math with Off-the-Shelf Models

(ends 8:30 AM)

Atlanta Orals 1B Theory: Online Algorithms, Dynamics & Complexity [7:30-8:30]

Orals 7:30-8:30

[7:30] The Power of a Random Sample in Online Algorithms

[7:30] The Stability of Online Algorithms in Performative Prediction

[7:30] Uniform-in-Time Weak Propagation of Chaos in Shallow Neural Networks

[7:30] Breakeven complexity: A new perspective on neural partial differential equation solvers

(ends 8:30 AM)

8:30 a.m.

Break:

Coffee & Snacks

(ends 9:00 AM)

Atlanta Poster Session 2 [8:30-11:30]

Posters 8:30-11:30

3D-PLOT-LLM: Part-Level Object Tokens for 3D Large Language Models

AbFlowNet: Optimizing Antibody-Antigen Binding Energy via Diffusion-GFlowNet Fusion

A Bitter Lesson for Data Filtering

Accuracy vs. Accuracy: Computational Tradeoffs Between Classification Rates and Utility

ACED-Bench: Evaluating How LLM Agents Acquire and Act on Causal Evidence

A Constrained Bi-level Optimization Framework for Constrained Preference-Based Reinforcement Learning

Actionable Hallucination Detection: Translating Latent Uncertainty into Agentic Critique

Action Images: End-to-End Policy Learning via Multiview Video Generation

Activation Functions Shape Token Synchronization in Stochastic Transformer Dynamics

ActWorld: From Explorable to Interactive World Model via Action-Aware Memory

Adaptive Mass-Segmented KV Compression for Long-Context Reasoning

Adaptive Power Iteration Method for Differentially Private PCA

AdERA: Adaptive Exponent Reuse for Lossless Allgather in Sharded MoE Training

A Doubly Smoothed Decentralized Stochastic Minimax Optimization Algorithm

AdpSplit: Error-Driven Adaptive Splitting for Faster Geometry Discovery in 3D Gaussian Splatting

Agentic Video Editing from Underspecified Requests

Agent-ToM: Learning to Monitor Autonomous LLM Agents via Theory-of-Mind Reasoning

Almost Sure Convergence of Linear Temporal Difference Learning with Arbitrary Features

Almost Sure Convergence Rates of Stochastic Approximation and Reinforcement Learning via a Poisson-Moreau Drift

A Margin Perspective on LoRA: Robustness to Catastrophic Forgetting and Adapter Merging (MaLoRA)

AME-TS: Anchored Mixture-of-Experts for Time Series Forecasting

AMPS: Adaptive Modality Preference Steering via Functional Entropy

Analytical Correction for Subsampling Bias in Drifting Models

A New Framework for Quantum Reinforcement Learning

A Semantic-Sampling Framework for Evaluating Calibration in Open-Ended Question Answering

A simple model of co-emergence of grid and place fields

A Single Neuron Is Sufficient to Bypass Safety Alignment in Large Language Models

A Theory on Flow Matching with Neural Networks

Attention Is More Than You Need: Spectral Redundancy in Multi-Head Routing

Attribute-Efficient Learning of Sparse Halfspaces with Constant Malicious Noise Rate

Augmented Lagrangian Method for Last-Iterate Convergence for Constrained MDPs

Automating ML for Science: Can Frontier Agents Climb Scientific Hills in the Wild?

AVO: Agentic Variation Operators for Autonomous Evolutionary Search

Beyond Bounded Variance: Variance-Reduced Normalized Methods for Nonconvex Optimization under Blum-Gladyshev Noise

Beyond Pairs: Your Language Model is Secretly Optimizing a Preference Graph

Beyond the Full Slate: Evaluating MNL Algorithms on All Slates

Bifurcation Models: Learning Set-Valued Solution Maps with Weight-Tied Dynamics

Can Entry-Wise Clipping Give Spectral Control of Stochastic Gradients?

Capturing LLM Capabilities via Evidence-Calibrated Query Clustering

CLUE: Correlated Latent Uncertainty for Single-Pass Deep Uncertainty Estimation

Community-Centered AI is Feasible and Beneficial for Impacted Communities

Comparing Transformers and Hybrid Models at the Token Level

Complementing DINO Features with Image Structure for Part Discovery

Composable Causality: A Toolkit for Systematic Time-Series Causal Discovery and Treatment-Effect Benchmarking

Constrained Decoding for Diffusion Language Models via Efficient Inference over Finite Automata

Continual Robot Learning via Language-Guided Skill Acquisition

Continuous Diffusion Scales Competitively with Discrete Diffusion for Language

Controllable Road Marking Generation

Convex Compositional Reasoning Models

CoTrek: Toward Scalable On-Policy Distillation for Long Chain-of-Thought Reasoning

CSI-TextBench: A Dataset and Benchmark for Language-Grounded Ambient Sensing Perception

CUDABeaver: Benchmarking LLM-Based Automated CUDA Debugging

Decentralized Aggregation of LLM Predictions via Wagering Mechanisms

Decision-Focused Learning in MDPs: An Occupancy Measure Approach

Denoising Implicit Variational Inference

Designing Kernel Surrogate Models for Multimodal Attribution

Diffeomoprhism-Informed 3D Gaussian Splattings via Screen-Space Optimal Transport (OT)

Differential Item Functioning as an Item-Level Diagnostic for LLM Benchmarks

Diffusion Language Models Can Approximate Optimal Infilling Lengths Implicitly

Discrete Stochastic Localization for Non-autoregressive Generation

DisSparse: Pipelined Top-$p$ Sparse Attention for Long-Context LLM Serving

Distilling Sequential Computation in Transformer Language Models

DivMoE: Fine-Grained MoE Upcycling via Cross-Domain Expert Composition

Do Joint Audio-Video Generation Models Understand Physics?

Do Latent-CoT Models Think Step-by-Step? A Mechanistic Study on Sequential Reasoning Tasks

Do multimodal models imagine electric sheep?

Do Robust LVLMs Hallucinate More? Uncovering Robustness-Induced Hallucination in Large Vision-Language Models

DRTriton: Large-Scale Synthetic Data Driven Reinforcement Learning for Triton Kernel Generation

DUIL: Deep Unsupervised Inverse Learning for in situ Macromolecular Morphology Identification

Dynamically Structured Diffusion Language Model Decoding via Bayesian Inference

Dynamic k-center clustering with lifetimes

Effective Multi-sensor Conditioning for Street-view Novel-view Synthesis

Efficient Algorithms For Fully Dynamic Bipartite Matching In Metric Spaces

Efficient LLM Adaptation with Forward-Only Passes

Elephant in the Fridge: Constant-Memory Frame Packing for Long Video Understanding

EMA-Nesterov: Stabilizing Nesterov's Lookahead for Accelerated Deep Learning Optimization

EpiPivot: Learning to Control the Simplex Method under Epistemic Uncertainty

Epistemic Infrastructures of Science in AI Era Should Rebalance Costs of Generation and Verification

Equivariant Force Field Calibration for Flow-based Protein Design

Ergodic Trajectory Design by Learned Pushforward Maps: Provable Coverage via Conditional Flow Matching

Evaluating Depth and Breadth in Test-Time Scaling for Compositional Visual Generation

Exact Unlearning via Quantized Sufficient Statistics

Exploiting Textual Semantics for Robust Cross-View Object Correspondence

Fast Accurate Quantum Monte Carlo without Metropolis Adjustment

Fast Diverse Nearest Neighbor Search

FASTER: Value-Guided Sampling for Fast RL

Fault Tolerant Coresets

FiedlerPrune: Connectivity-Preserving Cross-Layer Pruning for Large Language Models

FILOsofer: A TEE-Shielded Model Partitioning Framework based on Fisher Information-Guided LoRA Obfuscation

Finite-Resolution Decision Sufficiency for Linear Optimization

Finite Time Analysis of Risk-Sensitive RL via Noisy Power Iteration

Fixed Universal Transformers

Flash-KMeans: Fast and Memory-Efficient Exact K-Means

Flow Map Denoisers: Traversing the Distortion-Perception Plane for Inverse Problems

Flow Mismatching: Unsupervised Anomaly Detection via Velocity Discrepancies in Flow Matching Models

FlyAOC: Evaluating Agentic Ontology Curation of Drosophila Scientific Knowledge Bases

FOCUS: Benchmarking Retinal Model Generalization from Foundation Vision Encoders to Multimodal LLMs

ForceBody: Force-Paired Parametric Body Motion with Torque Uncertainty

From Matrix Inversion to Constraints: Provably Tighter Confidence Regions for Importance Weights in Label Shift

FrontierOR: Benchmarking LLMs' Capacity for Efficient Algorithm Design in Large-Scale Optimization

Frontier Task Synthesis Via Solution-Centric Evolution

Function graph transformers universally approximate operators between function spaces

FuRA: Full-Rank Parameter-Efficient Fine-Tuning with Spectral Preconditioning

GauS: Differentiable Scheduling Optimization via Gaussian Reparameterization

GenEnv: Difficulty-Aligned Co-Evolution Between LLM Agents and Environment Simulators

Generative OOD-regularized Model-based Policy Optimization

GeoFidelity-Bench: Evaluating Block-Conditioned Geographic Fidelity in Text-to-Image Street-View Generation

Geometrically Disentangling Concept Learning from the Language Modeling Loss

Ghosted Layers: Unconstrained Activation Alignment for Recovering Layer-Pruned LLMs

Golden Layers and Where to Find Them: Improved Knowledge Editing for Large Language Models Via Layer Gradient Analysis

Guiding Data Allocation for Robust Subpopulation Generalization

GUI-Libra: Data-Efficient Post-Training for Reliable Reasoning-and-Acting in Native GUI Agents

G-Zero: Self-Play for Open-Ended Generation from Zero Data

Handwriting decoding as a challenging motor task for EEG Foundation Models

HearSayBench: Can LLMs Navigate from Abstract Human Rights to Lived Lives?

H-Flow: Self-supervised Human Scene Flow via Physics-inspired Joint Multi-modal Learning

Hierarchical Denoising For Multi-Step Visual Reasoning

HorizonComposer: Spatiotemporally Consistent Driving Video Editing with Enriched Traffic Semantics

How well behaved is finite dimensional Diffusion Maps embedding?

Hypothesis generation and updating in large language models

Improving constraint-based discovery with robust propagation and LLM priors

In-Context Multi-Operator Learning with DeepOSets

Inferential Theory of Learning as a Framework for Test-Time Computation in Foundation Models

Inline Critic Steers Image Editing

Inside Emergence: Structure-Behaviour Gaps in Language Model Training

Interpreting Latent Protein Language Model Features with Geometric Annotations

Intra-Option Fitted Q-Evaluation: Evaluating Hierarchical Policies from Non-Hierarchical Data

Intrinsic Riemannian Cross-covariance for Manifold-valued Random Objects

Inverted Detection and Control in Steering Vectors

I-PTC: Interactive Programmatic Tool Calling for Stateful Tool-Augmented Agents

Is Decentralized LLM Agent RL Robust to Heterogeneity? An Asymmetric Tale

JARVIS-Bench: Benchmarking Personal Intelligence Agents on Long-Horizon Real-User Daily Traces

K-PWM: Control-Oriented Structured World Models under Partial Observation

KVBuffer: IO-aware Serving for Linear Attention

Language-Induced Priors for Domain Adaptation

Learning Biological Hierarchies in Single-Cell Foundation Models

Learning Evidence Highlighting for Frozen LLMs

Learning in Context, Guided by Choice: A Reward-Free Paradigm for Reinforcement Learning with Transformers

Learning Rate Decay Can Exponentially Accelerate SGD for Global Optimization of Nonconvex Functions

Learning Rate Transfer in Normalized Transformers

Learning Task-Centric World Models from Visual Foundations

Learning to Follow In-Context Watermark Instructions via Self-Distillation

Learning Unbiased Permutations via Flow Matching

Learning What Not to Impute: An Uncertainty-Aware Diffusion Framework for Meaningful Missingness

Learning What to Predict: Downstream-Guided Task Design for Continued Pretraining

Learning Where and What to Restore for Composite Image Restoration

Less Structure is More: Minimal Representations for Supervised Learning

Leveraging Latent Visual Reasoning in Silence

Leviathan: Decoupling Input and Output Representations in Language Models

Linguistic Trajectory Encoding for Efficient Long-Horizon Spatial Memory in Embodied Agents

Localizing Concepts in Visual Autoregressive Models

LogicDirector: Enforcing Temporal Composition in Text-to-Video Generation

Long-Lived AI Agents Age Too: They Quietly Decay After Deployment

Lost on Campus: Evaluating Embodied Spatial Reasoning of Vision-Language Models in the Wild

Magnitude-preserving Layers Enable Efficient GANs

Masked Diffusion Language Models are Strong and Steerable Text-Based World Models for Agentic RL

MetaCluster: Enabling Deep Compression of Kolmogorov-Arnold Network

Minimax Optimal Kernel Two-sample Testing in Sub-quadratic Time

MinSteer: Minimal-Pair Steering via Two-Stage Cached Continuation

MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI

MoBiQuant: Mixture-of-Bits Quantization for Token-Adaptive Any-Precision LLM

Mode-Controlled Policy Optimization: A Geometry-Aware Recipe for LLM Post-Training

Model-Free Assessment of Simulator Fidelity via Quantile Curves

Monotone Inclusion Approach to Weakly Monotone Discrete-Time Finite-Horizon Mean-Field Games

MPCI-Bench: A Benchmark for Multimodal Pairwise Contextual Integrity Privacy Evaluation of Language Model Agents

Multi-Stage Planning from Single-Stage Data: Reinforcement Learning Helps Composition but Requires Anchoring

Multi-Turn RL Makes Small Language Model Competitive for Optimization Modeling

Nearest-Neighbor Radii under Dependent Sampling

Near-optimal Explainable $k$-means Clustering under $\ell_p$ Norm

Neural-Corrected Operator Learning for Homogenization and Inverse Design

NeuronEye: Query-Guided Visual Concept Activation for Vision-Language Reasoning

Non-Colliding Biometric Identities for Digital Entities: Geometry, Capacity, and Million-Scale Virtual Identity Provisioning

Not Every Rubric Teaches Equally: Policy-Aware Rubric Rewards for RLVR

OmniGF: A Dual-Branch Vision-Language Framework for Unified Gaze Following

On Inherent Privacy of Posterior Sampling: A Unified R\'enyi-Divergence Framework

Online Allocation with Differential Privacy

On Neural Scaling Laws for Weather Emulation through Continual Training

On-Policy Hindsight Distillation for Early Risk Prediction

On the Complexity of Offline Reinforcement Learning with Q*-Approximation and Partial Coverage

OpenBrain: An Auditable Generated-Label Release for Whole-Brain MRI Parcellation

Optimal Byzantine-resilient Federated Learning with User-level Differential Privacy

Optimizer-Induced Mode Connectivity: From AdamW to Muon

Pacing Branch Parallelism in LLM Serving

Partition-Aware Unlearning for Removing Spurious Correlations in Large Vision-Language Models

pCoMole: Pareto-Constrained Molecule Editing with Discrete Flows

PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models

Personalized and Collaborative Online LQR via Thompson Sampling

Persuasive Prediction via Decision Calibration

PixelDense: Dense Prediction as Representation Alignment for Pixel Diffusion

POME: Post Optimization Model Edit via Muon-Style Projection

Position: AI-Agent Pricing Should Become More Outcome-Dependent: An Economic Perspective

Position: Reconciling Open Access with Owner Control in AI Model Distribution Deserves More Research Effort

Precise Debugging Benchmark: Is Your Model Debugging or Regenerating?

Predictive–Generative Drift Decomposition for Speech Enhancement and Separation

Primal Generation, Dual Judgment: Self-Training from Test-Time Scaling

Private Adaptive Covariance Estimation via Gaussian Graphical Models

Progress-Aware Distillation for Mitigating Stagnation in Small Language Model Agents

Projection Learning: A Principled Way to Overcome Memorization in Distribution Learning

ProjKAN: Model Compression via KAN Projections to Bridge the Hypothesis and Capacity Gaps

Proximal Difference-in-Differences for Long-Term Causal Learning under Confounding and Outcome Drift

Prune, Don’t Rebuild: Efficiently Tuning $\alpha$-Reachable Graphs for Nearest Neighbor Search

RA$^2$: Retain-Anchored Attraction Preserves Forget-Adjacent Utility in LLM Unlearning

Rate-Constrained Edge Metadata for Sender–Receiver Generative Video Super-Resolution

Rational Tuning of LLM Cascades via Probabilistic Modeling

RATS! Patches Talk Through Registers: Emergent Parts in Register Attention Transformers

Recreating Video Arenas via Automated Preference Scoring

RefineTok: Scale-Wise Tokenization for Progressive Visual Refinement

RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer

Reinforcement Learning for Diffusion LLMs with Entropy-Guided Step Selection and Stepwise Advantages

Reinforcing Multimodal Reasoning Against Visual Degradation

Reject, Resample, Repeat: Understanding Parallel Reasoning in Language Model Inference

Representation Fréchet Loss for Visual Generation

Rethinking LoRA Initialization for Robust Asymmetric Learning Rates

Rethinking On-Policy Self-Distillation for Thinking Models

Rethinking Structured Generation: Can Graph-Based Reasoning Resolve Ambiguity?

Revealing the Gap in Human and VLM Scene Perception through Counterfactual Semantic Saliency

Reviving Stale Updates: Data-Free Knowledge Distillation for Asynchronous Federated Learning

RL-Inf: Tracking Non-local Training Data Influence for Online Reinforcement Learning

Robust and Efficient Finetuning of Vision Foundation Models via Implicit Ensembling

Robust Diffusion Models via Divergence-Induced Weighted Denoising

Robust Graph Diffusion Model

Robust Nash Alignment under Preference Uncertainty

ROCKET: Residual-Oriented Multi-Layer Alignment for Spatially-Aware Vision-Language-Action Models

RustMizan: A Compilable, Contamination-Aware Benchmarking Framework for Rust Vulnerabilities

S$^{2}$-PINN: Stochastic Separable Physics-Informed Neural Networks

SAFTAC: Simulation-Augmented Fine-Tuning of Open-Source LLMs for Analog Circuit Design

Sample complexity of stochastic optimization with integer variables

Scalable Derivative Gaussian Processes via Exact Gradient Reduction

SCULPT: Advancing Masked Discrete Diffusion for High-Resolution Image Synthesis.

SEISMOS: A Statistical Signal Detection Framework for Semantic Chunking

Sharpness of Minima in Deep Matrix Factorization

SHRED: Retain-Set-Free Unlearning via Self-Distillation with Logit Demotion

SiliciclasticReservoirs: A Million-Reservoir Dataset and Flow-Matching Foundation Model for 3D Siliciclastic Reservoir Generation

SiliconBench: Speed, Memory, and Fidelity for LLM Inference on Apple Silicon

SimplexUQ: An Evaluation Framework and Benchmark for Conformal Uncertainty on Simplex-Valued Predictions

SlopCodeBench: Benchmarking How Coding Agents Degrade Over Long-Horizon Iterative Tasks

SLVR: Structured Latent Visual Reasoning via Human-like Reasoning Flows

Smooth Flow Matching for Synthesizing Functional Data

Sort, Partition, Randomize: Optimal Binary Hypothesis Testing under Local Differential Privacy

SparDA: Sparse Decoupled Attention for Efficient Long-Context LLM Inference

Spectral-Sphere-Constrained Hyper-Connections

Speed Predictions for Online Energy-Efficient Scheduling

Spend Only What You Need: Defect-Aware Residual Coverage for Efficient Multi-Agent Reasoning

StarCraft Motion: A Dataset for Agent Simulation in Adversarial and Partially Observable Scenarios

StaRPO: Stability-Augmented Reinforcement Policy Optimization

Statistical Query Lower Bounds for Smoothed Agnostic Learning

Steering Vectors as a Training Signal in LLM Post-Training

StereoPep: Do Molecular Models Understand Stereochemistry? A Benchmark on Synthetic Diastereomeric Peptides

StomataBench: Measuring Taxonomic Generalization in Stomatal Detection

Stop Quantum Machine Learning; do AI-for-Quantum instead

Stop Using Plausibility as the Criterion for Explainable AI

Strategic Decision Focused Learning

Structure-agnostic Causal Representation Learning

SyncWorld: Visual Calibration Enables World Models as Zero-Shot Simulators

TEMPO: Temporal Enforcement via Mode-Separated Policy Optimization for Trustworthy LLM Backtesting

Testable Learning of General Halfspaces under Massart Noise

Test-time Risk Adaptation with Mixture of Agents

The Master Key Hypothesis: Unlocking Cross-Model Capability Transfer via Linear Subspace Alignment

The Missing Positional Story in LLMs: A Case Study of Shift-Invariant Attention

There are Levels to It: Red Teaming LLMs with Hierarchical Reinforcement Learning

The Score Kalman Filter

The Stability of Online Algorithms in Performative Prediction

TKCAM: Text and Keyframe to Camera Trajectory Generation

Topology-Reinforced Swin Transformer for Medical Image Analysis

TorchUMM: A Unified Multimodal Model Codebase for Evaluation, Analysis, and Post-training

To See Far, Look Close: Evolutionary Forecasting for Long-term Time Series

Towards Differentially Private Reinforcement Learning with General Function Approximation

Towards Hyperparameter Transfer for Differentially Private Optimization

Towards Reliable LLM Evaluation: Correcting the Winner’s Curse in Adaptive Benchmarking

TraceDx: Criticality-Weighted Atomic Facts as a Training Signal for Sequential Clinical Diagnosis

Tracing Agentic Failure from the Flow of Success

TrajWiki: Source-Grounded Memory Trajectories for Long-Horizon Dialogue Agents

Transfer Learning Through Conditional Quantile Matching

Transforming Image Editors into Video Editors

Tri-Prompting: Controllable Video Generation with Scene, Subject, and Motion Prompts

TritonTune: LLM-Guided Multi-Agent Optimization of GPU Kernel Configurations

UMAS: System-Level Uncertainty Quantification for Multi-Agent LLM Systems

Uncertainty Aware SURE Transfer Learning for Classification Problems

Uncertainty-Guided Reward Labeling for Reinforcement Learning under Limited Feedback

Understanding Double Descent through Universal Compression

Understanding Graph Self-Supervised Pre-training under Distribution Shifts: A Scaling Law Perspective

UniPath: Adaptive Coordination of Understanding and Generation for Unified Multimodal Reasoning

UniReFP: Robust Unified Fingerprinting for Vision Models against Cross-Task Repurposing Attacks

Useful Memories Become Faulty When Continuously Updated by LLMs

Value Entanglement: Conflation Between Different Kinds of Good In (Some) Large Language Models

Variational Wasserstein Model on Riemannian Manifolds for Image Segmentation

Verifiers in the Loop: Decoding Time Verification for Code Translation

ViT-AdaLA: Adapting Vision Transformers with Linear Attention

What should post-training optimize? A test-time scaling law perspective

When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning

When Dynamics Shift, Robust Task Inference Wins: Offline Imitation Learning with Behavior Foundation Models Revisited

When Policies Cannot Be Retrained: A Unified Closed-Form View of Post-Training Steering in Offline Reinforcement Learning

Where You Backpropagate Matters: Token Hypothesis for Memory-Efficient Fine-Tuning

Why Cancer cfDNA Models Fail on Chronic Disease: A Geometric Information-Theoretic Bound and Its Architectural Implications

World Action Verifier: Self-Improving World Models via Forward-Inverse Asymmetry

(ends 11:30 AM)

11 a.m.

Remarks:

Welcome Remarks

(ends 11:30 AM)

11:30 a.m.

Reception:

Welcome Reception

(ends 12:30 PM)

FRI 11 DEC

1 a.m.

Atlanta Invited Talk:

Stephanie Jegelka

Stefanie Jegelka

(ends 2:00 AM)

2 a.m.

Break:

Coffee & Snacks

(ends 2:30 AM)

Atlanta Poster Session 3 [2:00-5:00]

Posters 2:00-5:00

Accelerating Inference of Discrete Autoregressive Normalizing Flows by Selective Jacobi Decoding

Activation Steering of Video Generation Models via Reduced-Order Linear Optimal Control

ActQuant: Sub-4-bit Action-Guided Quantization for Vision-Language-Action Models

Adaptive Target-Charging with Privacy Filters and Individual Accounting

AdaState: Self-Evolving Anchors for Streaming Video Generation

A Distribution Mapping Approach to Counterfactually Fair Reinforcement Learning

Agentic AI-Empowered Dynamic Survey Framework

Agent MechSuits : Mechanistic Subspace Safety Steering for Multi-Turn CLI Agents

AgentOdyssey: Open-Ended Long-Horizon Text Game Generation for Test-Time Continual Learning Agents

A Locality-Aware Surrogate for Natural-Gradient Descent in Quantum Optimization

A Mechanistic Interpretability Study of an Astronomical Foundation Model

AndroidReality: How Far Are Mobile Agents from the Real World?

A Provably Convergent and Practical Algorithm for Gromov–Wasserstein Optimal Transport

Ares: Adaptive Reasoning Effort Selection for Efficient LLM Agents

Argus: A Cross-Regime Benchmark for the Transferability of Uncertainty Quantification in Computer-Use Agents

Asymmetric Factorization for Low-Rank PSD Learning: When Is the Relaxation Exact ?

A Systematic Analysis of Out-of-Distribution Detection Under Representation and Training Paradigm Shifts

Attend Locally, Remember Linearly: Linear Attention as Cross-Frame Memory for Autoregressive Video Diffusion

Autoregressive Diffusion World Models for Off-Policy Evaluation of LLM Agents

Axiomatic World Modeling for Physics Reasoning

Beyond Incremental Beam Search for Compositional Explanations of Neurons

Beyond Interpretability: When, Why, and How Sparse Autoencoders Enable Label-Free Visual Steering

Beyond Semantic Similarity: Rethinking Retrieval for Agentic Search via Direct Corpus Interaction

Beyond Task Success: Probing Cognitive Primitives in Web Agents

Beyond the Trial-and-Error Loop: Hybrid Projection and Automated Tuning for Distributed Training

Boosting Text-to-Image Diffusion Models via Core Token Attention-Based Seed Selection

Bottleneck Benchmark: Sequence Compression Under Controlled Difficulty and Fixed Latent Size

BrainTRACE: Tracing Longitudinal, Multimodal, and Volumetric Evidence in Brain MRI Clinical Reasoning

Breaking the Bias Barrier in Concave Multi-Objective Reinforcement Learning

Breaking the Grid: Distance-Guided Reinforcement Learning in Large Discrete Action Spaces

Brute-Force Jailbreaks and Codon-Aware Watermarking for DNA Foundation Models

Budgeted Multi-Source Counterfactual Annotation for Off-Policy Evaluation

Budgeting Discretion: Theory and Evidence on Street-Level Decision-Making

Bypassing PC1 Makes SAEs More Reproducible

CAFE: Causally-Guided Automated Feature Engineering with Multi-Agent Reinforcement Learning

Calibrating Agentic LLMs for Clinical Prediction

Can Bits Seal Language?

Categorical Bayes filtering for computational phenotyping in adaptive learning

Causal Benchmarks for Multimodal AI Should Measure Categorical Difference, Not Capability Gap

CellularS2-Bench: A Staged, Evidence-Grounded Benchmark for Cellular Network Security Reasoning

Chain-of-Generation: Progressive Latent Diffusion for Text-Guided Molecular Design

CHoRD: Coordinating Scheduling and Data Placement for Efficient Deep Neural Network Inference on Chiplet-Based GPUs

CineOrchestra: Unified Entity-Centric Conditioning for Cinematic Video Generation

Class Adaptive Conformal Training

Clean-Label Poisoning for Gradient-Boosted Decision Trees

Closing the Reflection Gap: A Free Calibration Bonus for Agentic RL

Coarse-to-Fine 3D MRI Reconstruction via Resolution-Agnostic Neural Operators

Coding with "Enemy": Can Human Developers Detect AI Agent Sabotage?

Co-Evolving Interpolants and Flows via Path-Flow Alignment

Cognitive bias benchmarks should incorporate insights from ecological rationality

Computationally sufficient statistics for Ising models

Concurrent Image Understanding and Generation: Self-Correcting Coupled Markov Jump Process

Conditional Multi-Event Temporal Grounding in Long-Form Video

Contrastive Discovery: Open-Ended Scientific Discovery over Competing Explanations

Controllable Molecular Generative Foundation Models

Convergence of Near-Linear Width ReLU Networks with Unbalanced Initialization

COPRA: Conditional Parameter Adaptation with Reinforcement Learning for Video Anomaly Detection

CORP: Closed-Form One-shot Representation-Preserving Structured Pruning for Transformers

Corrective Diffusion Language Models

COSAC: Counterfactual Credit Assignment in Sequential Cooperative Teams

Cost Efficient Fairness Audit Under Partial Feedback

Coverage-Based Calibration for Post-Training Quantization via Weighted Maximum Coverage over Outlier Channels

Cross-Distribution Generalization in Longitudinal Behavioral Data Through Frozen Coherence Constraints

Cross-Fitting for Neural Posterior Estimation

CURe: Conservative Unlearning with Soft-Gating Regularization for Offline Reinforcement Learning

DARLING: Detection Augmented Reinforcement Learning with Non-Stationary Guarantees

DARPAN: Controllability-Aware Residual Filtering for Pretrained Physical Control

Decoupled Descent: Exact Test Error Tracking Via Approximate Message Passing

Deep Reasoning in General Purpose Agents via Structured Meta-Cognition

DeEscalWild: A Real-World Benchmark for Automated De-Escalation Training with SLMs

DEFT: Disentanglement-Enhanced Fine-Tuning for EEG Foundation Models

DeltaPrompts: Escaping the Zero-Delta Trap in Multimodal Distillation

Denoising Time Matters: Diverse Generation in Diffusion Language Models

Diagnosing and Repairing Citation Failures in Generative Engine Optimization

DiffeoMorph: Learning to Morph 3D Shapes Using Differentiable Agent-Based Simulations

Differentiable Systematic Resampling for Variational Sequential Monte Carlo

Diff-Instruct with Diffused Reward: Towards Principled One-step Generator RL

Diffusion Model's Generalization Can Be Characterized by Inductive Biases toward a Data-Dependent Ridge Manifold

DiRotQ: Rotation-Aware Quantization for 4-bit Diffusion Transformers

DistDebug-Bench: Can LLM Agents Diagnose Bugs in Distributed Systems?

Distributionally Robust Token Optimization in RLHF

Does Sparse Connectivity Improve Generalization? Convolutional Networks Below the Edge of Stability

Do Graph Neural Networks Learn Generalizable Algorithms for Clustering Graphs?

Do Pathology Foundation Models Encode Disease Progression? A Pseudotime Analysis of Visual Representations

DORA: A Scalable Asynchronous Reinforcement Learning System for Language Model Training

Do Semantic Distance Tests Actually Predict Creativity in Large Language Models?

DriveMind: Mind-Evolving Belief Tracking for Closed-Loop Autonomous Driving

DSAD: Dynamic Soft Anisotropic Diagrams for Reduced-Order Video Representation

Dual-Granularity Learning for Regression with Continuous Noisy Labels

Dynamic Expert Sharing: Decoupling Memory from Parallelism in Mixture-of-Experts Diffusion LLMs

Dynamic Quadtree Tokenization for Autoregressive PDE Forecasting

Dynamic Tokenization via Reinforcement Patching: End-to-end Training and Zero-shot Transfer

Early Prediction of Future Behavioral Strategy from Process Traces

EgoMo3R: Joint Egocentric Motion and Scene Reconstruction

Elastic Spectral State Space Models for Train-Once Budgeted Inference

Emergence, Retention and Mitigation of Ill-conditioning due to Basis Lifting in KANs

Entropy Dynamics of Agent Reinforcement Learning

Evaluating Compositional Generalization in Transformers: The Role of Composition Equivalence and Module Coverage

Event-Grounded Sparse Autoencoders for Vision-Language-Action Policies

Exposing the Illusion of Erasure in Knowledge Editing for LLMs

Finite-Sample Performance of Gradient Descent in Logistic Regression with Gaussian Design

Firefly: Illuminating Verified Real-world Tool Call Data Generation

Fixed-Size Active Statistical Inference

Flow-Based Conformal Predictive Distributions

Flow Matching for Offline Reinforcement Learning with Discrete Actions

ForecastCompass: Guiding Agentic Forecasting with Adaptive Factor Memory

Free Draft-and-Verification: Toward Lossless Parallel Decoding for Diffusion Large Language Models

From Collapse to Improvement: Statistical Perspectives on the Evolutionary Dynamics of Iterative Training on Contaminated Sources

From Detection to Understanding — A Multi-Task Dataset for Traffic Anomaly Reasoning

GAIA: Geometry-Adaptive Operator Learning for Forward and Inverse Problems

Generalized Robust Adaptive-Bandwidth Multi-View Manifold Learning in High Dimensions with Noise

Generalizing Test-time Compute-optimal Scaling as an Optimizable Graph

Generation-for-Understanding with Structured Action Scripts for Embodied Multimodal Learning

Geometry-Centered 3D Latent World Models for Growing Surfaces

GlucoFM-Bench: Benchmarking Time-Series Foundation Models for Blood Glucose Forecasting

GraphIP–Bench: How Hard Is It to Steal a Graph Neural Network, and Can We Stop It?

Graph Label Alignment: A Diagnostic Atlas for Graph Classification

GRAPHLCP: Structure-Aware Localized Conformal Prediction on Graphs

GRNAgent: A Multimodal Graph Reasoning Agent for Gene Regulatory Network Inference

Hallucination in World Models is Predictable and Preventable

HDCS: Hierarchy Discovery and Critic Shaping for Reinforcement Learning with Automaton Specification

HEART: Hyperspherical Embedding Alignment via Kent-Representation Traversal in Diffusion Models

HERO: Improving the Reliability and Sensitivity of Generative Model Evaluation Using Historical Data

Hide-and-Seek in Trajectories: Discovering Failure Signals for VLA Runtime Monitoring

HilbertGen-3D:Hilbert--Multifractal Conditioning for Topology-Aware 3D Generation

Inference and estimation with unidentifiable latent treatment effects

Inference and Uncertainty Quantification for Streaming $r$-PCA

InfoSFT: Learn More and Forget Less with Information-Aware Token Weighting

Invertible Logits Transformation for Accuracy-Preserving Post-Hoc Uncertainty Calibration

Is $\sqrt{d}$ Separation Necessary for Gradient EM to Learn Gaussian Mixtures in High Dimensions?

Is Backpropagation Optimal? When Synthetic Gradients Improve Sample Efficiency

Joint Confounder Selection for Causal Mediation Analysis in High Dimensions

Justitia: Fair and Efficient Scheduling of Task-parallel LLM Agents with Selective Pampering

K9-Bench: Evaluating Multimodal LLMs on Canine-Centric Videos

Latent Q-Barrier Shielding for Safe In-Context Reinforcement Learning

Leaner transformers can easily learn to cluster

Lean Refactor: Multi-Objective Controllable Proof Optimization via Agentic Strategy Search

Learning a Maximum Entropy Model for Visual Textures using Diffusion

Learning Cultural Vectors for Cross-Cultural Generation

Learning How to Cube

Learning Latency-Aware Orchestration for Multi-Agent Systems

Learning Sparse Compositional Functions with Norm-Constrained Neural Networks

Learning to Trigger: Reinforcement Learning at the Large Hadron Collider

Learning What to Remember: Test-Time Training via Context Distillation

Learning When to Denoise: Optimizing Asynchronous Schedules for Latent Diffusion

Learning When Visual Context Matters for Mouse Behavior Analysis

LensCT: Fine-Grained AI-Involved Text Detection via Temporal-Hierarchical Tomograms of LLM Internals

Lie Generator Networks for Nonlinear Partial Differential Equations

LLMs can construct powerful representations and streamline sample-efficient supervised learning

LoRASpace: A Pool-Wide Shared Substrate for Static and Dynamic Multi-LoRA Composition

Loss is Not Behavior: A Unified Output-Space Analysis of Gradient-Based Machine Unlearning

Loyalty Capture: Reporting Relationships and Structural Sycophancy in Frontier AI Models

Masked Sobolev Training for Feasibility-Reliable Optimization Proxies

Massively Parallel Exact Inference for Hawkes Processes

MAST: Label-Efficient, Robust, and Generalizable Sound Detection for Biodiversity Monitoring via Masked Audio Pretraining and Self-Training

Measuring and Mitigating the Distributional Gap Between Real and Simulated User Behaviors

Meta$^n$: Recursive Self-Improvement through Emergent Depth

MetaPI: Constructing Prompt Injection Benchmarks from Any Agent Benchmarks

Mind the Gap: Information Disadvantage as a Learning Signal in Cooperative MARL

MindVLM: Neural-Grounded Visual Captioning via Subject-Aware Semantic Evidence Selection

MIRROR: Manifold Ideal Reference ReconstructOR for AI-Generated Image Detection

MixUni: Joint Multi-Property Prediction for Chemical Mixtures with Physics-Informed Heads

Model-Adaptive Tool Necessity Reveals the Knowing-Doing Gap in LLM Tool Use

Models Recall What They Violate: Constraint Adherence in Multi-Turn LLM Ideation

MotorSense: A Video-EMG Dataset of Motoric Representations for Action Understanding

Multi-Head Recurrent Memory Agents

Multi-Objective Causal Bandits: Minimal Intervention Space and Policy-Level Learning

Multi-Objective Reinforcement Learning Using Routed Ensembles and Trajectory Attribution

Multi-Token Residual Prediction

Neural Dual Bounds: Valid-by-Construction JGLP Warm-Starts for MAP and Constrained MAP

Neural Statistical Functions

Not all uncertainty is alike: volatility, stochasticity, and exploration

Not Suppressing or Purifying: Backdoor Containment via Expert Quarantine and Shutdown in LLMs

OdysSim: Building Foundation Models for Human Behavior Simulation

One-Bit Clustering for Two Component Sub-Gaussian Mixture Models

Online Bayesian Calibration under Gradual and Abrupt System Changes

On the Efficiency of Structured Pruning in Small Language Model Pretraining

On the Error Correcting Effects of Stochasticity in Discrete Diffusion

On the Impact of Side-Information in Bandit Learning: More is Not Always Merrier

On Worst-Case Guarantees for Graph-Based Nearest-Neighbor Search

Optimal and Efficient Contextual Combinatorial Semi-bandits with General Function Approximation

Optimal Linear Regression Without a Variance

Oracle-Robust Online Alignment for Large Language Models

Pan-FM: A Pan-Organ Foundation Model with Saliency-Guided Masking for Missing Robustness

PanoWorld: Geometry-Consistent Panoramic Video World Modeling

PCBSchemaGen: Reward-Guided LLM Code Synthesis for Printed Circuit Boards (PCB) Schematic Design with Structured Verification

PEEK: Context Map as an Orientation Cache for Long-Context LLM Agents

Persona Vectors: Monitoring and Controlling Character Traits in Language Models

Posterior Alternative Calibration in Ambiguous Inverse Problems

Powering Up Zeroth-Order Training via Subspace Gradient Orthogonalization

Precision As You Need: Stochastic Computing Is a Dense Adaptive Quantizer

PRISM: Primitive Routing via In-context Skill Mixing for Lifelong VLA

Procedural Memory Distillation: Online Reflection for Self-Improving Language Models

PRO: Enabling Precise and Robust Text Watermark for Open-Source LLMs

Prompt-Driven Exploration

Proposing Better Rollouts for On-Policy Distillation

PuppetGait: Generalizing Gait Recognition via 3D Body-Aligned LVM Features

Quotient-Categorical Representations for Bellman-Compatible Average-Reward Distributional Reinforcement Learning

RADAR: Text-Guided Medical Image Segmentation via Residual Aggregation and Dense Alignment Representations

Rare-Tail Statistics For Learning Biased Gaussian Halfspaces with Label Noise

RAVEL: Rare Concept Generation and Editing via Graph-driven Relational Guidance

Reading, Not Thinking: Bridging the Modality Gap When Text Becomes Pixels

Reasoning Gestures: LLM-Inferred Communicative Functions for Co-Speech Gesture Generation

RefDecoder: Enhancing Visual Generation with Conditional Video Decoding

Relaxed On-Policy Distillation: Selective Credit Allocation for Scaling Reasoning Efficiently

Reliable Abstention under Adversarial Injections: Lower Bounds and New Upper Bounds

ReSCUE: Re-translation with Sentence Commitment for Unsegmented Long-Form Simultaneous Sign Language Translation

Response Time Enhances Alignment with Heterogeneous Preferences

Rethinking Contrastive Targets in Radiology Language-Image Pretraining

Rethinking Geometric Depth in Monocular 3D Object Detection: A Projection-Consistent Reformulation

Rethinking Ratio-Based Trust Regions for Policy Optimization in Multi-Agent Reinforcement Learning

Rethinking SAE Evaluation: An Atomic Interpretable Unit Framework Reveals Hidden Polysemanticity and Redundancy

Rethinking Semantic ID Construction for Generative Recommendation: SimHash with Parallel Decoding and Semantic Alignment

Revisiting the Generic Transformer: Deconstructing a Strong Baseline for Time Series Foundation Models

Reward-Conditioned Reinforcement Learning

Reward Shaping to Improve Language Model Query Generation

RigidFormer: Learning Rigid Dynamics using Transformers

RILA: A Radar-Native Structured Interface from Sparse mmWave Point Clouds to Large Language Models

RoboWits: Unexpected Challenges for Robotic Creative Problem Solving

Robust Surrogate Modeling for Explainable Graph Neural Networks

RPP: A Certified Poisoned-Sample Detection Framework for Backdoor Attacks under Dataset Imbalance

Safe-PG-LQR: Provably Safe, Convergent and Optimal Model-Free Linear Quadratic Regulator with Hard Constraints

SAGE: Mitigating Long-Horizon Reasoning Biases via Topological Guidance

SAIR: Cost-Efficient Multi-Stage ML Pipeline Autoscaling via In-Context Reinforcement Learning

Scale-mixture Langevin sampling in the subspaces of recurrent cortical circuit dynamics

Scale Where It Matters: Training-Free Localized Scaling for Diffusion Models

SceneBind: Binding What and Where Across Vision, Audio, and Language

SchemeArena: Factorized Stress Testing of Scheming in LLM Agents

ScreenShot: A Foundation Model for Few-Shot Combination Drug Screening

Semantic Routing: Exploring Multi-Layer LLM Feature Weighting for Diffusion Transformers

SGD Provably Prioritizes a Shortcut Spurious Feature in the XOR Model

ShadowTransfer: A Geographic Transfer Benchmark for Overhead Shadow Detection

Shellsort as Multi-Scale Relaxation: Learning Spectrum-Matched Gap Schedules

SkillRL: Evolving Agents via Recursive Skill-Augmented Reinforcement Learning

SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks

Sliced Inner Product Gromov–Wasserstein Distances

SMI: Semantic Medical ID for Hierarchy-Aware Concept Representation

Smoothed Score Queries and the Complexity of Sampling

SnapAudit: Active Auditing of Differentially Private In-Context Learning via Snapshot-Based Simulation

Soft Token Alignment for Cross-Lingual Reasoning

Sparse Reward Subsystem in Large Language Models

SPATIALEPIBENCH: Benchmarking Spatial Information and Epidemic Priors in Forecasting

Spectral-Spatial Interpretation

Spin-Weighted Spherical Harmonics Enable Complete and Scalable E(3)-Equivariant Networks

Split-RL: Local Conflict Resolution in Reinforcement Learning

Stability and Diversity of Networked Self-Consuming Generative Ecosystems

StepBack: Step-Level Error-Localized Resampling for Efficient Test-Time Reasoning

Stepwise Penalization for Length-Efficient Chain-of-Thought Reasoning

STRAND: Sequence-Conditioned Transport for Single-Cell Perturbations

Strategic Decision Support for AI Agents

Stress-Testing Neural Network Verifiers with Provably Robust Instances

(Strongly) Replicable Distribution Testers imply High Probability Distribution Testers

Structurally Separated DAG Learning with Multi-Scale Normalized Closure

Structure-Adaptive Estimation of Heterogeneous Treatment Effects with Kernel Methods

Swarm Shepherd: Securing Multi-Agent Ecosystems Against Persistent Latent Compromise

Swimba: Switch Mamba Model Scales State Space Models

Synthesis Through Simulation: Generating Coherent Enterprise Data via Scalable Agent-System Interaction

Synthetic Tasks for Training AutoResearch

Synthon Contrastive Learning for Synthesizable 3D Molecule Generation

TabDLM: Free-Form Tabular Data Generation via Joint Numerical–Language Diffusion

TabWorld: A World-Modeling Foundation Model for Tabular Generation

Tail Wags the Model: Generalization and Membership Privacy Trade-offs of Sharpness-Aware Minimization

Taking the Road Less Scheduled with Adaptive Polyak Steps

TargetSage: Identifying Therapeutic Target Genes with Interpretable and Robust LLM Reasoning

The Best-Laid SCHEMEs: Coordinated Sabotage and Monitoring in Multi-Agent Systems

The Capability Frontier of GRPO

The Geometry of Generative Latents: Navigability and Invertibility

The Many Faces of On-Policy Distillation: Pitfalls, Mechanisms, and Fixes

Theory-Scale Auto-Formalization of Logics for Computer Science

The parameters in weight-sparse transformers are interpretable

The Stability of Data Exchange in Competitive Markets

The Type Theory of Stationary MDPs: Rare Events and Uncertainty Quantification

TIER: Trajectory-Invariant Execution Rewards for Multi-Step Tool Composition

Token Inflation: How Dishonest Providers Can Overcharge for Large Language Model Usage

TopoGraphRAG-Bench: Evaluating Multimodal GraphRAG on Layout-Grounded Evidence Reasoning

Towards a Universal Causal Reasoner

Towards Matrix-Parallel and Feature-Scalable 3D Gaussian Splatting Rendering on GPUs

Towards Mitigating Deceptive Safety Alignment in Large Reasoning Models

Trimming the Long-Tail of Visual World Modeling Evaluation

Trust Region Continual Learning as an Implicit Meta-Learner

Trust the Direction, Search the Step: Zero-and-First-Order Methods for LLM Fine-Tuning

Two is better than one: designing heterogeneous scales in binary rating systems

UltraDiff:Transferring High-Fidelity Priors to Compressed Latent Spaces for High-resolution Image Generation

Understanding Reasoning from Pretraining to Post-Training: Chess as a Controlled Testbed

UtoMe: Observation-Uncertainty-Guided Token Merging for Weather Foundation Models

Value-Aware Stochastic KV Cache Eviction for Reasoning Models

Variance-Optimal State Resampling for Reinforcement Learning with Verifiable Rewards

VAR-Q: Tuning-free KV Cache Quantization for Visual Autoregressive Image and Video Generation

VICO: Visual Environments Co-Evolving for Vision-Language Model Reasoning

Vision-Language Grounding as Bidirectional Concept Correspondence

Vision-Language Models Should Commit to a Cézannian Specification

Visual Robustness and Neural Alignment in a Shared Foraging Task: The Mouse vs. AI Benchmark

Wasserstein residuals: Learning Gradient Flows from Population Dynamics

What if Agents Could Imagine? Reinforcing Open-Vocabulary HOI Comprehension through Generation

When and How Much to Imagine: Adaptive Test-Time Scaling with World Models for Visual Spatial Reasoning

When the Same Coefficients Reach Different Places: Asymmetric Realizability in Transplanting Tokenizers across Large Language Models

Why Pass@k Optimization Can Degrade Pass@1: Prompt Interference in LLM Post-Training

World Tracing: Pixel-Aligned Geometry Beyond the Visible

XBRIDGE: Entity-Grounded Latent Bridge for Heterogeneous LLM Communication

XMNoise2Clean: Cross-Modal Denoising Under Sparse Data

You Only Need Minimal RLVR Training: Extrapolating LLMs via Rank-1 Trajectories

Your Benchmark Is an Empirical Measure Over Difficulty

Your Hypergradient is Skewed: Antithetic Neumann Estimation for Bilevel Optimization

AI and Data Science for Her: Training the Next Generation of High School Girls in Nigeria

The Voice-Native Classroom: Multilingual AI Co-Teachers for Global ML Education

DTCH Digital Turcologist Bootcamp: Evidence-First, Evaluation-First AI/ML for Under-Resourced Historical Turkic Languages and Cultural Heritage Data

The 100x Challenge: Compression-First Research Pipeline for Bandwidth-Constrained Regions

FATE-AI: Futures of Agriculture, Technology, and Ecology in Artificial Intelligence

Wikair (Student-Led Nonprofit), a Global, Youth-Led, Interdisciplinary AI Education Program

(ends 5:00 AM)

4:30 a.m.

Break:

Retail Lunch

(ends 6:30 AM)

7:30 a.m.

Atlanta Orals 2A Generative Modeling: Diffusion, Visual & 4D Generation [7:30-8:30]

Orals 7:30-8:30

[7:30] Generative Modeling via Drifting

[7:30] Representation Fréchet Loss for Visual Generation

[7:30] Query Lower Bounds for Diffusion Sampling

[7:30] Trellis4D: Complex 4D Mesh Generation

(ends 8:30 AM)

Atlanta Orals 2B Trustworthy Vision: Explanation, Forensics & Identity [7:30-8:30]

Orals 7:30-8:30

[7:30] DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging

[7:30] SALART-VQA: Diagnosing Whether VLMs Understand Salient Artifacts in Generated Images

[7:30] Video Forensic Self-Descriptions: Leveraging Temporally Distributed Forensic Microstructures for Zero-Shot Detection of AI-Generated Videos

[7:30] Non-Colliding Biometric Identities for Digital Entities: Geometry, Capacity, and Million-Scale Virtual Identity Provisioning

(ends 8:30 AM)

8:30 a.m.

Break:

Coffee & Snacks

(ends 9:00 AM)

Atlanta Poster Session 4 [8:30-11:30]

Posters 8:30-11:30

\$OneMillion-Bench: How Far are Language Agents from Human Experts?

3D Point Splatting for mmWave Radar Novel View Synthesis

Acting without Knowing: Planning, Prediction, and Transfer Dissociate in Interactive Visual Physics

Actor-Accelerated Policy Dual Averaging for Reinforcement Learning in Continuous Action Spaces

Act, Validate, Adapt: Closing the Causal Discovery-Control Loop

A Curvature Phase Transition Governs Coherence Penalty Efficiency Against Feature Absorption in SAEs

AgentCollabBench: Diagnosing When Good Agents Make Bad Collaborators

AI Governance Should Prioritize Control and Knowledge Boundaries Over Limiting Intelligence

A Latent-Load Framework for Reliability Analysis and Intervention Design in LLM Pipelines

AlgoPilot: Cross-Paradigm Reasoning in Language Models via Strategy Selection and Guidance

Aligning Forest and Trees in Images & Long Captions for Visually Grounded Understanding

Alignment Is Not Enough for Safe Medical LLM Evaluation

All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow

A low-rank decoder bottleneck bounds reliability in foundation-model perturbation prediction

AmbientFM: A Foundation Model for Ambient Sensing

AnaDiffusion: Anatomically Compositional Latent Diffusion for Controllable 3D Brain MRI Generation

AnyHand: A Large-Scale Synthetic Dataset for RGB(-D) Hand Pose Estimation

Approximate Matrix–Vectors Under a Bounded $\ell_1$ Assumption and Applications to Kernel Matrices

ArcMark: Distortion-Free Multi-Byte LLM Watermark via Optimal Transport

A Steerable Deep Network for Model-Free Diffusion MRI Registration

A Theory of Adversary-Directed Online Learning

Attention Sinks and Outliers in Attention Residuals

Auditing Evidence Framing at NeurIPS: A Decade-Scale Study of Accepted Papers

A Unified Spectral Theory of Multimodal Losses

A Unified Uncertainty Representation for Graph Neural Networks via Doubly-Spectral Stochastic Expansion

Backdoor Purification for LoRA-Tuned LLMs via Null-Space Projection

Back to Blackwell: Closing the Loop on Intransitivity in Multi-Objective Preference Fine-Tuning

Bayesian Additive Distribution Regression

Benchmarks as Measurement Instruments: Quantifying Signal and Noise for More Efficient AI Evaluations Under Distribution Shift

Bernini: Latent Semantic Planning for Video Diffusion

Beyond 3 Million Tokens: A Multi-Modal Foundation Model for Full-Resolution Heliophysics

Beyond Copy-Paste: How Well Do Subject-Driven Video Models Understand Their Subjects?

Beyond Empirical Support: Structured Outlier Generation via Sinkhorn Optimal Transport

Beyond LoRA vs. Full Fine-Tuning: Gradient-Guided Optimizer Routing for LLM Adaptation

Beyond Marginal Coverage: Efficient Localized Conformal Prediction via Residual Rank Calibration

Beyond Pessimism: Offline Learning in KL-regularized Games

Beyond Scalar Distances: Semantic Attribute Gradients from Frozen MLLMs for Visual Embeddings

Bridging Textual Profiles and Latent User Embeddings for Personalization

Budgeted Quotient-Residual Guidance for Frozen Pocket-Conditioned Molecular Diffusion

BuresTomFlow: Bures-Geometric Flow Matching for Posterior Quantum Tomography

Calibration without labels in multiple testing

Cardinality-Decomposed Loss for Heterogeneous GNNs

Causal Effects with Unobserved Unit Types in Interacting Human–AI Systems

Causal Multi-Task Demand Learning

CCDiff: Inverse Canonical Correlation Analysis for Discovering Visual Differences in Natural Language

ChanSFormer: A Channel Agnostic Vision Transformer for Multi-Channel Cell Painting Images

CipherFlow: Hardware-Aware Compiler Framework for Low-Latency Hybrid Secure Inference

CLAMP: A Sim-to-Real Benchmark for Closed-Loop Kinematic Pose Estimation and Assembly Reasoning

Clustering with Weak Distance Oracles

CMPQ: Compensatory Quantization via Input-Aware Hessian Damping

CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs

Coherent Routing in Decision Trees: Phase-Interference Learning for Interpretable Tabular Prediction

CoMMa: Contribution-Aware Medical Multi-Agents for Decentralized Oncology Decision Support

Compute Aligned Training: Optimizing for Test Time Inference

Conceal, Reconstruct, Jailbreak: Exploiting the Reconstruction--Concealment Tradeoff in MLLMs

Concept Modulation Models: A Unified Framework for Identifiability and Extrapolation

Conformal Prediction with Paraphrase-Aware Scoring for LLM Uncertainty Quantification

Constrained Code Generation with Discrete Diffusion

Constraint-Aware Influence Estimation in Deep Constrained Learning

Consumer Search and Social Learning in Agentic Markets

Controllable and Content Based Recommendations

Coreset-Induced Conditional Velocity Flow Matching

Counterfactual Distillation: Internalizing Reflective Experience into LLM Agent

Covariate-Adjusted Deep Causal Learning for Heterogeneous Panel Data Models

CupOFMoCA: Coupled Objective-Guided Discrete Flows for Molecular Conjugate Assembly

Curriculum proof repair with learned counterexamples

DD-CAM: Minimal Sufficient Explanations for Vision Models Using Delta Debugging

DEBATE: A Large-Scale Benchmark for Evaluating Opinion Dynamics in Role-Playing LLM Agents

Decision Focused Scenario Learning for Contextual Stochastic Programming

Demystifying Classifier-Free Guidance for Auto-Regressive Image Generation

Dense Flow from Adaptive Correspondence

DICE: Decoupling Capability from Intervention Necessity in LLM Tutoring

DiffATS: Diffusion in Aligned Tensor Space

DoFP-Aligned Lookup Tables for Real-Time Polarization Demosaicking

DopplerWild: A Doppler Dataset and Benchmark for Human Kinematic Understanding in the Wild

Do Thinking Tokens Help with Safety?

ECHO: Terminal Agents Learn World Models for Free

Efficient Test-time Adaptation through Candidate Verification and Divergence Shifts

Elastic Representations via Hyperbolic Geometry

Embedding Security Properties into AI-Enabled Cyber-Physical Systems

Enabling approximate joint sampling in diffusion LMs

Ensemble Selective Classification

Entropy Distribution as a Fingerprint for Hallucinations in Generative Models

Escaping Reasoning Basins: Basin-Aware Search for Inference-Time LLM Reasoning

Evaluating an Evaluation: Membership Inference Attacks as Machine Unlearning Diagnostics

FaceProbe: Recovering HDR Environment Map via Masked Diffusion and Physical Preference

Fair Division of Work in Collaborative Mean Estimation via Bargaining

Fairness Failure Modes of Multimodal LLMs

Fast Rates for Offline Contextual Bandits with Forward-KL Regularization under Single-Policy Concentrability

FlashEvolve: Accelerating Agent Self-Evolution with Asynchronous Stage Orchestration

FlashFFN: Multi-Head Decomposition Enables I/O-Aware Feed-Forward Network

Flash-SD-KDE: Accelerating SD-KDE with Tensor Cores

FlowBank: Query-Adaptive Agentic Workflows Optimization through Precompute-and-Reuse

Free energy Estimation on Any State Space

From Nodes to Pixels: Topological and Structural Two-View Graph Imaging

From Zero to Hero: Training-Free Custom Concept Spawning in World Models

GazeFlow: From Human Gaze Behavior to Generative Egocentric Gaze Prediction

Generalized Smooth Stochastic Variational Inequalities: Almost Sure Convergence and Convergence Rates

Generalizing Action-Conditioned Latent World Models with Video Model Rewards

Generative Conformal Prediction with Optimized Coverage Allocation

Geometry-Aware Score-Repellent Monte Carlo

Geometry over Density: Few-Shot Cross-Domain OOD Detection

Global Optimality for Constrained Exploration via Penalty Regularization

GlucoFM: A Dual-Stream Foundation Model for Continuous Glucose Monitoring

Graph Anomaly Detection as Dynamical Transport: Training-Free Scoring via Empirical Bayes

Graph Sparse Sampling: Breaking the Curse of the Horizon in Continuous MDP Planning

Grouped Adaptive Head Mixing for Personalized Multi-Task Federated Reinforcement Learning

GroupMemBench: Benchmarking LLM Agent Memory in Multi-Party Conversations

Harder the Task, Sparser the Representation: Sparsity as a Learning Signature of Capability in LLMs

HESTIA: A Hessian-Guided Differentiable Quantization-Aware Training Framework for Extremely Low-Bit LLMs

Hierarchical Graph Alignment for Cross-Modal 3D Scene Grounding

HierFlow: Hierarchical Coupled Dual-Space Search for Automatic Agentic Workflow Generation

High-dimensional Gaussian Graphical Model Testing for Long-Memory Time Series

How does RL Post-training Induce Skill Composition? A Case Study on Countdown

Hyper Input Convex Neural Networks for Shape Constrained Learning and Optimal Transport

Imperfect Influence, Reliable Rankings: A Theory of TRAK for Data Attribution

Improved Leverage Score Sampling for Constrained Active Linear Regression

Inference Time Nash Alignment

Informed Posterior Sampling: More Efficient Online Learning with Few Offline Demonstrations in Average-Reward MDPs

INFUSER: Influence-Guided Self-Evolution Improves Reasoning

Introspection Tools Help LLMs Understand and Control Themselves

Inverse Reinforcement Learning with Just Classification and a Few Regressions

Iterative Gumbel Planning for Continuous Control

Jet-Long: Efficient Long-Context Extension with Dynamic Bifocal RoPE

JODA: Composable Joint Dynamics for Articulated Objects

Knowing Without Saying: How Contextual Evidence Survives but Fails to Surface in Transformers

Know Your Task, Learn It Right: Task-Aware Optimistic Value Learning for Multi-Task Multi-Agent Reinforcement Learning

Language Models Need Sleep

Layerwise Progressive Freezing: A Training Scaffold for Depth-Scalable Binary Networks

LCD$^3$: Layout-Conditioned Diffusion for Dataset Distillation in Object Detection

Learnable Low-Rank Polynomial Sketch for Effective Linear Attention

Learning a Unified Cross-Model Semantic Dictionary via Gated Bottleneck Sparse Autoencoders

Learning from Trials and Errors: Reflective Test-Time Planning for Embodied LLMs

Learning Preference Representations for Preference-Conditioned Image Generation

Learning Recoverable Neural Networks against Weight Corruption via Simple Zero-Sum Projection

Learning When to Think: Adaptive Internal Computation for Reinforcement Learning

Less Decoder is More Encoder: Geometric Representation Learning from Novel View Synthesis

Limits and Potential of Score-Based Data Valuation: Redundancy, Complementarity, and Non-Monotonicity

LLM-ACES: Closed-Loop Discovery of Dynamic Systems with LLM-Guided Adaptive Search

LLMs Improving LLMs: Agentic Discovery for Test-Time Scaling

Low-Rank Hierarchical Merging for Efficient Long-to-Short Reasoning

Majority Bit-Aware Watermarking for Large Language Models

MammoGPS: A Benchmark for Visual Grounding, Perception, and Spatial Reasoning in Mammography

MaPPO: Maximum a Posteriori Preference Optimization with Prior Knowledge

Matrix-Free Stochastic Training of Low-Rank Spectral Graph Learning via Randomized Adaptive Spectral Estimation

Measuring and Strengthening Behavioral Suppression in Language Models

Mechanistic Critics for Sample-Efficient NPU Design-Space Exploration

Mecha-nudges for Machines

MedVIGIL: Evaluating Trustworthy Medical VLMs Under Broken Visual Evidence

MEMAUDIT: An Exact Package-Oracle Evaluation Protocol for Budgeted Long-Term LLM Memory Writing

MiCo: Microstructure-Consistent Flow Matching for Diffusion MRI Angular Super-Resolution

Minimally Invasive Steering of Language Models

Minimax-Optimal Transformer Classification for Functional Data with Dense-Sparse Phase Transition

Mitigating Label Bias with Interpretable Rubric Embeddings

ML-assisted Randomization Tests for A/B Experiments

Model-based Bootstrap of Controlled Markov Chains

Model Cascades with Provable Per-Class Quality

More Is Not More: What Matters for Diversity in LLM Opinions？

Motion Cues from Image-based Point Tracking for LiDAR Scene Flow Estimation

MOVEBENCH: A Benchmark for Global-Scale Wildlife Movement Forecasting

MT-JailBench: A Modular Benchmark for Understanding Multi-Turn Jailbreak Attacks

Multi-agent Collaboration with State Management

Nearly Optimal Attention Coresets

Near-Optimal Last-Iterate Convergence for Zero-Sum Games with Bandit Feedback and Opponent Actions

No Coin Left Behind: Maximizing Strategic Surplus Against No-Regret Dynamics

Offline Preference-Based Trajectory Evaluation

OLA-Place: Cross-Modal Place Recognition without Global Descriptors

One-Shot Generative Flows: Existence and Obstructions

One-Shot Private Confidence Regions via Resampling

Online Finetuning Decision Transformers with Pure RL Gradients

Online Maximization of Non-Decomposable Test and Population Utilities

On Lipschitz Explosion in Deep Neural Networks with Normalization: Consequences for Optimization and Robustness

On the Parallel Optimality of Exponentiated Gradient Descent

On the Primacy Bias in RLVR Training

On the Token Value Inequality in Efficient Reasoning

OpenWebRL: Demystifying Online Multi-turn Reinforcement Learning for Visual Web Agents

OpInf-LLM: Parametric PDE Solving with LLMs via Operator Inference

Optimal Contextual Pricing under Agnostic Non-Lipschitz Demand

Optimal Rates for Pure $\varepsilon$-Differentially Private Stochastic Convex Optimization with Heavy Tails

Optimal Recalibration of an Online Predictor

Optimizing Retraining Schedules via Learning Curves

Order-based structure learning for zero-inflated count data under data heterogeneity

Persistent-Transient Policy Evaluation for Markov Chains via Minimal Peripheral Quotients

PhysGraphNet: Physical-State Scene Graphs via Latent Graph Reasoning and Counterfactual Supervision

Physics-Conditioned Video Diffusion with Kinematic Priors for Fusion Capsule Polishing

PIGRAM: An Interpretable Patch--Motif Interaction Grammar for Protein--Nucleic-Acid Recognition

PIVOT: A Unified Agentic Framework for Streaming Long-Video Understanding

PLACE: Patch-Level Agnostic Concept Extraction

Policy-DRIFT: Dynamic Reward-Informed Flow Trajectory Steering

Position: LLM Privacy Requires a Lifecycle-Wide Approach

Positive-Unlabeled Preference Optimization For Chest X-ray Report Generation

Predicting and improving test-time scaling laws via reward tail-guided search

Predicting Plasticity in Deep Continual Learning: A Theoretical Perspective

Prompt–Activation Duality: Improving Activation Steering via Attention-Level Interventions

Quantum Speedup of Multi-armed Bandits at Scale by Tackling Memory Decoherence

Qubrio: High-Performance Quantum Compilation via Multi-Agent LLM Collaboration

Query Lower Bounds for Diffusion Sampling

RAD-TFM: Robust and Domain-Adapted Tabular Foundation Models

RAHF: Reward-Amplified Human Feedback for Closed-Loop Policy Fine-Tuning

Rank Is Not Capacity: Spectral Occupancy for Latent Graph Models

RAVEN-Bench: A Paired EO-IR Video QA Benchmark for Aerial Multimodal Understanding

Reading Positional Coupling in Transformers with Diffusion Scores

RECIPE: Procedural Planning via Grounding in Instructional Video

RecoverBench: A Systematic Benchmark for Error Recovery in Robotic Manipulation

ReCoVer: Resilient LLM Pre-Training System via Fault-Tolerant Collective and Versatile Workload

RefineAny3D: Depth Refinement as Semantic Alignment for Monocular 3D Detection

Regret–Oracle Complexity Tradeoffs in Agnostic Online Learning

Reinforcement learning enhanced flow matching for reference-based tumor generation on unpaired CT images

REINS: Learning Inertia-Induced Geometry for Physics-Consistent Motion Representation in Clinical Gait Phenotyping

Rep2Text: Decoding Full Text from a Single LLM Token Representation

RepFusion: Leveraging Multimodal Priors for Denoising in Representation Space

RepoLaunch: Automating Build and Management of Code Repositories across Languages and Platforms

Residual-Autoregressive Context for 3D Gaussian Splatting Compression

Rethinking the Readout: Unlocking Video Backbones for AI-Generated Video Detection

RICE-PO: Turning Retrieval Interactions into Credit Signals for Reasoning Agents

Risk-Averse Online POMDP Planning via CVaR of the Immediate Cost with Performance Guarantees

Row-Private Symmetric Cone Programming: Scale-Efficient Algorithm and Lower Bound

RSPO: Reasoning-Supervised Policy Optimization for Long-Tail Autonomous Driving

RxGS: Receiver-Generalizable 3D Gaussian Splatting for Radio-Frequency Data Synthesis

Safe in Its Own Words: Self-Guided Safety Alignment for Multimodal Reasoning Models

SAGE: Evidence-First Biomarker Discovery through Multi-Agent Reasoning

Sample Complexity of Linear Regression under Random-Location Coordinate Corruptions

SarcBench: A Bilingual Benchmark for Contextual Sarcasm Understanding, Response, and Generation

Say the Same, Act Differently: Text-Orthogonal Action Subspaces in Reasoning Vision-Language-Action Models

Scaling Arbitrary Architectures and Optimizers with Automatic Parameterization

SCDM: Scalable Causal Discovery in Nonlinear Temporal Systems with Meta-Learning

SceneAligner: 3D-Grounded Floorplan Localization in the Wild

SceneFactory: GPU-Accelerated Multi-Agent Driving Simulation with Physics-Based Vehicle Dynamics

ScreenSearch: Uncertainty-Aware OS Exploration

Searching Videos as Trees: Self-Correcting Agents for Grounded Long Video QA

SEED: Self-Speculative Decoding via Implicit Encoder–Decoder

See it to Place it: Evolving Macro Placements with Vision Language Models

Self Driving Datasets: From 20 Million Papers to Nuanced Biomedical Knowledge at Scale

SemanticDialect: Semantic-Aware Mixed-Format Quantization for Video Diffusion Transformers

ShadowBench: Exposing Lexical Anchoring and the Illusion of Forgetting in Large Language Models

Signature Approach for Contextual Bandits with Nonlinear and Path-dependent Rewards

SLOT-IR: Learning Disentangled Slot Representations for Infrared Spectral Unmixing

sMMC-22M: A Context-Aware Dataset and Benchmark for Single-Cell Spatial Transcriptomics

Smoothed Elicitation Complexity for Approximate $\Gamma$-calibration of Discrete Classification Tasks

SPANUQ: Span-Level Uncertainty Quantification for Large Language Model Generation

Sparse Attention as a Range Searching Problem: Towards an Inference-Efficient Index for KV Cache

Sparse Expansion Utility: Identifying and Routing to Pivotal Steps in LLM Reasoning Chains

Sparse Internal Control of Language Models

Speakeasy: Auditing Cheap Signals in Peer Prediction

SpecDrop: Parameter-Free Category-Conditioned Routing for Modular Specialization

SpecForge: Agent-Oriented Code Documentation Optimization via Multi-Frontier Tree Search

SpecHop: Continuous Speculation for Accelerating Multi-Hop Retrieval Agents

Spectrally Parameterized Neural Inverse Reconstruction

Spectral Stratification of Semantic Abstraction in Vision-Language Models

Stochastic Approximation Approach for Decentralized Optimization on Time Varying Random Networks

Stop Mechanizing Reform Heuristics as Scientific Quality Filters in AI Review

Strong Teacher Not Needed? On Distillation in LLM Pretraining

Structure Over Scale: Learning Visual Reasoning from Pedagogical Video

StyleStream 2.0: Fast and Controllable Streaming Voice Style Conversion

Subliminal Transfer of Unsafe Behaviors in AI Agent Distillation

Synthetic Worlds for Temporal Evaluation and Knowledge Updating in LLMs

TAC: Timestamped Audio Captioning

Tcell: Mitigating Harmful Fine-tuning for Large Language Models via Gradient Alignment

Temporal Island Sparse Autoencoders for Interpreting Clinical Time-Series Models

TensorCommitments: A Lightweight Verifiable Inference for Large Language Models

Testing and Estimation of Contextual Generalized Thurstone Models

\texttt{FEROM}: Frontier Endogenous Reveal-Order Marginal Policy Optimization for Masked Diffusion LMs

The Complexity Kink: LLM Rubric Instruments for Causal Inference on Code Generation Reliability

The Cost of Absolute Position: A Spread-Expressivity Tradeoff for Additive Positional Encodings

The Invisible Hand of Physics: When Video Diffusion Models Know More Than They Show

Theoretical Limits of Language Model Alignment

Theory on Attention Dynamics for Out-of-Distribution In-Context Learning

The Power of a Random Sample in Online Algorithms

The Power of Menus in Dynamic Pricing: Near Optimal Regret and Equivalence Results

The Prestige: Benchmarking Cognitive Visual Reasoning using Magic Tricks

The Reasoning Boundary Paradox: How Reinforcement Learning Constrains Language Models

The Value of Being Wrong: Self-Mined Visual In-Context Learning from Errors

To trust or not to trust: Attention-based Trust Management for LLM Multi-Agent Systems

Toward in Silico Strain Evaluation: A Multimodal Surrogate for Fermentation Dynamics with Metabolic Graph Pretraining

Towards Reconstructing Geographically Diverse Architecture with 3D Foundation Models

Training Agent to Scale Inference-Time Reasoning

Train on the Sphere, Deploy on the Hill: Closed-Form-Anchored Surrogates for Real-Terrain Boundary-Integral Equations

Trajectory Planning without Trajectory Data: A Manifold-Guided Approach

Transolver-GMsFEM: A Hybrid Framework for High-Contrast Multiscale PDEs on Irregular Grids

Two-Fidelity Best-Action Identification for Stochastic Minimax Tree

Ultra Fast PDE Solving via Physics Guided Few-step Diffusion

Understanding the Effects of Neuron Dominance in Deep Reinforcement Learning

UniVL: Unified Vision-Language Embedding for Spatially Grounded Contextual Image Generation

Unlearning Diffusion Policies via Relative Fisher Forgetting

Unsupervised Concept Discovery with Dirichlet Concept Diffusion Models

Variational Inference via Entropic Transport Descent

VeriContest: A Competitive-Programming Benchmark for Verifiable Code Generation

Verify0: Can AI Agents Build Formally Verified Software Repositories?

VerifyThisBench: Joint Evaluation of Code, Specifications, and Proof

VFIG: Vectorizing Complex Figures in SVG with Vision-Language Models

Vision to Geometry: 3D Spatial Memory for Sequential Embodied MLLM Reasoning and Exploration

VolFill: Single-View Amodal 3D Scene Reconstruction with Volumetric Flow Matching

Watermarking Without Standards Is Not AI Governance

When and Why is Optimistic Multiplicative Weights Slow? The Geometry of Energy Dissipation

When Are Compositional Problems Learnable from Verifiable Rewards?

When Attribution Patching Lies: Diagnosis and a Second-Order Correction

When Expert Disagreement Hurts: Auditing Prestige-Sensitive Revision in LLM Decision Pipelines

When Form Changes but Logic Doesn’t: Building Logic-invariant LLMs through Structures

When Simulation Lies: A Sim-to-Real Benchmark and Domain-Randomized RL Recipe for Tool-Use Agents

When, Where, What: Structural Guarantees for Travel Time Prediction on Temporal Graphs

Where Tabular Foundation Models Falter on Genetic Data: Datasets That Expose and Provide a Path to Address the Gap

Why Transformer-Based Language Models Need Explicit Mechanisms of Cognitive Control

AI and Data Science for Her: Training the Next Generation of High School Girls in Nigeria

The Voice-Native Classroom: Multilingual AI Co-Teachers for Global ML Education

DTCH Digital Turcologist Bootcamp: Evidence-First, Evaluation-First AI/ML for Under-Resourced Historical Turkic Languages and Cultural Heritage Data

The 100x Challenge: Compression-First Research Pipeline for Bandwidth-Constrained Regions

FATE-AI: Futures of Agriculture, Technology, and Ecology in Artificial Intelligence

Wikair (Student-Led Nonprofit), a Global, Youth-Led, Interdisciplinary AI Education Program

(ends 11:30 AM)

11 a.m.

SAT 12 DEC

1 a.m.

Atlanta Invited Talk:

Power to the People? Why AI Data Centers Face Growing Resistance, and How We Can Do Better

Sasha Luccioni

(ends 2:00 AM)

2 a.m.

Break:

Coffee & Snacks

(ends 2:30 AM)

Atlanta Poster Session 5 [2:00-5:00]

Posters 2:00-5:00

$\chi$-Bench: Can AI Agents Automate End-to-End, Long-Horizon, Policy-Rich Healthcare Workflows?

$\gamma$-weakly $\theta$-up-concavity: A Unified Framework for Non-Convex Optimization Beyond DR-Submodular and OSS Functions

A$^3$Bench: A Benchmark for Compositional Reasoning over Aggressive Interactions in Videos

Absolute State-wise Constrained Policy Optimization: High-Probability State-wise Constraints Satisfaction

Achieving adaptivity and optimality for multi-armed bandits using Exponential-Kullback Leibler Maillard Sampling

A Control-Theoretic Approximation to Predictive Coding Dynamics

Action Chunking Proximal Policy Optimization with Feedback Correction

Action-Level Behavior Policy Optimization for Variance-Reduced Policy Gradients

Active Learning for Conditional Generative Compressed Sensing

AdaAlloc: Adaptive Visual Token Allocation for Long-Video Question Answering

AdaPaD: Adaptive Parallel Deflation for PEFT with Self-Correcting Rank Discovery

Adapting Actively on the Fly: Relevance-Guided Online Meta-Learning with Latent Concepts for Geospatial Discovery

Adaptive Calibration in Non-Stationary Environments

Adaptive Delayed-Update Cyclic Algorithm for Variational Inequalities

Adaptive Test Case Discovery for LLM-Assisted Decision Making in High-Stakes Domains

AdShot: Benchmarking Multimodal Large Language Models for Video Advertisement Clipping

A Few Teacher Steps Go a Long Way: Cost-Efficient On-Policy Data Augmentation for Agent Post-Training

AgentArk: Distilling Multi-Agent Intelligence into a Single LLM Agent

AI-Assisted Classification under Correlation Neglect and Trust

AI Safety Evaluations Need More Human-AI Experiments

Algorithmic Impact Reveals the Hidden Structure of Alignment

Alignment Needs 'Cognitive Control': On The Role of Regularization in LLM Alignment

Allocate Marginal Reviews to Borderline Papers Using LLM Comparative Ranking

An Embarrassingly Simple Graph Heuristic Reveals Shortcut-Solvable Benchmarks for Sequential Recommendation

Approaching Effective Merging in Model Embedding Space

Approximation Algorithms for GPU Pricing under Finite Capacity

A Single-Sample Polylogarithmic Regret Bound for Nonstationary Online Linear Programming

A Training-Free Video Moment Retrieval Framework via Selection from Multiple Visual Prompts

Attention as In-Context Empirical Bayes: A Two-Stage View via Particle Dynamics

Attention-Based Pretraining for Unsupervised Amortized Causal Discovery

Audible World Models: Spatially Aware Sound Generation for 3D Worlds

Autonomous Driving Research Requires a Community-Driven Data Paradigm

A World Model of Radiologist Reading for Medical Image Representation Learning

B$^3$-PWL: GPU-Batched Branch-and-Bound for Piecewise-Linear Optimization with SOS2 Constraints

Be CARE-ful with Text-to-SQL Benchmarks

BehaviorBench: Modeling Real-World User Decisions from Behavioral Traces

Benchmark Recovery Does Not Certify a Frozen Tool-Trigger Contract After Quantization

Beyond Confidence: Rethinking Self-Assessments for Performance Prediction in LLMs

Beyond Thinking: Imagining in 360$^\circ$ for Humanoid Visual Search

Beyond Trajectory Matching: Reflow with Marginal Distribution Alignment

Beyond What Seems Necessary: Hidden Gains from Scaling Training-Time Reasoning Length under Outcome Supervision

BiMoGen: Bidirectional Motion-Text Generation via Unified Masked Discrete Diffusion

BLEND: Balancing Personalization vs. Generalization in Federated Vision–Language Models

Bridging Modalities, Spanning Time: Structured Memory for Ultra-Long Agentic Video Reasoning

Bridging Simulation and Reality: Geometry and Decision Alignment for Autonomous Driving

Capturing In-Context Learning Dynamics with Task Operators

CASPIAN: Online Detection and Attribution of Cascade Attacks in LLM Multi-Agent Systems via Cross-Channel Causal Monitoring

Causal Discovery from Unseen Environments

CFC26: Building Evaluations for Deployment in Sonar-Based Fish Counting

Characterizing Underrepresentation in Generalizing Causal Survival Estimates

Classification at the Edge of Stability: Unifying Self-Stabilization and Convergence Rates

CLR-voyance : Reinforcing Open-Ended Reasoning for Inpatient Clinical Decision Support with Outcome-Aware Rubrics

CogBench: Evaluating Cognitive-Level Control in LLM Question Generation

ColdDDI: Evaluating Knowledge Utilization in Cold-Start Drug-Drug Interaction Prediction

Compositional Policy Optimization with Language Models

Compositional Reasoning in Language Models under Reinforcement Learning Post-Training

Computational Dynamic Mechanism Design

Computer Science Conferences Should Require Nonrepudiable Experimental Results

Concept-Localized Generative Representations

Conditional Optimal Bridge for Riemannian Activation Steering

Conformal Selective Acting: Anytime-Valid Risk Control for RLVR-Trained LLMs

Contextualized Evaluation of Vision Language Models through Dynamic Interviews

Contrastive Nonmyopic Objective Cost-Tradeoff Acquisition for Longitudinal Data

Corrupted Plans, Clean Traces: What Planning-Execution Decoupling Reveals About CoT Monitoring

Counterfactual Predictive State Representations: The Intrinsic Dimension of Partial-Information Games

Coupled Integral PINN for Discontinuity

CRANE: Constrained Reasoning Injection for Code Agents via Nullspace Editing

CSO-LLM: Class Subspace Orthogonalization for Post-Training Backdoor Detection and Trigger Inversion in LLMs

CurveRL: Principled Distribution-Aware Context Reweighting for LLM Reasoning

CVTA: Cross-Variable Temporal Attention for Multivariate Irregular Time Series Prediction

DDBench: A Benchmark for Agentic Debugging on Distributed Systems

Debiasing Sketched Ridge Regression: A Functional Estimation Perspective

Decompose the Distillation: Interpretable Single-Pass Guidance for Diffusion Models

DeepVoting: Learning and Improving Voting Rules with Fine-Tuning

DeltaFugue: Orchestrating Spatial and Associative Memory for Algorithmic Length Generalization

Demystifying Numerical Errors in LLM Inference: Achieving Reproducible Inference for Mission-Critical Tasks with HEAL

Deployment-Time Online Imitation Learning from Corrective Demonstrations

Depth Exploration for LLM Decoding

DictLLM: Post-training Compression of Large Language Model with Dictionary Kernels

Differentiable Bit-Widths: Co-optimizing Pruning and Quantization via SVD for Ultra-Efficient LLM Compression

Differentially Private Sparse Reward Estimation with Preference Feedback

Do Deep Networks Forget Initialization? A Forgetting-Time View of Practical Inductive Bias

DOME: Drift-Adaptive On-Policy Motion Erasure in Video Diffusion Transformers

Don't Always Pick the Highest-Performing Model: An Information Theoretic View of LLM Ensemble Selection

Don't Deploy Fine-Tuned Genomic Foundation Models Without Privacy Evaluation: Reconstruction Vulnerability Is Unpredictable Without Empirical Measurement

DPrivBench: Benchmarking LLMs' Reasoning for Differential Privacy

DropKV: Decoupling Residual-Output Perturbation for Near-Optimal KV-Cache Eviction

Efficient Analytic Uncertainty Quantification for Multimodal Regression

Efficient and Transferable Agentic Knowledge Graph RAG via Reinforcement Learning

Efficient Collaborative LLM Fine-Tuning over Heterogeneous Mobile Devices via Many Backbones to One Side-Network Tuning

Efficiently Representing Algorithms With Chain-of-Thought Transformers

Efficient Multi-Source Prompt Adaptation for Cross-Domain Open-Vocabulary Learning

Efficient Neural Field Learning via Adaptive Coverage and Focused Sampling

Efficient Transferable Optimal Transport via Min-Sliced Transport Plans

ELSA3D: Elastic Semantic Anchoring for Unified 3D Understanding and Generation

Embedding Foundation Model Predictions in Discrete-Choice Models with Structural Guarantees

Embodied AI Cannot Scale Without Open-Source Distributed Geometric Optimization Backends for Life-Scale Egocentric Data

Enabling Denoising Score Matching Type Training for Manifold Diffusion Model via Momentum and Splitting

Estimating the expected output of wide random MLPs more efficiently than sampling

EVALUATION CARDS: An Interpretive Layer for AI Evaluation Reporting

Even Sharper Bounds for Transductive Learning and Its Applications

Evolution Fine-Tuning: Learning to Discover Across 371 Optimization Tasks

Exploiting Fine-Tuning Structures to Improve Adversarial Transferability on Downstream SAM

Exploring MLLM-Diffusion Information Transfer with MetaCanvas

Factor Augmented High-Dimensional SGD

Factored Generative Models through Mechanism Diversity

FactorizedHMR: A Hybrid Framework for Video Human Mesh Recovery

FairTune Market: A Fair and Trustworthy Marketplace for Fine-Tuned LLMs via Posted-Price & Proper-Scoring Mechanisms

Fast Approximate $\ell_p$ Chamfer Distance via Lopsided Embeddings and Structured JL

FedCF: Fair Federated Conformal Prediction

Few-Step Diffusion Language Models via Trajectory Self-Distillation

Finite-Sample Convergence in Networked Average Reward MARL: Decentralization Pitfalls and Entropy Remedies

Flow Matching Reinforcement Learning via SDE Inference

Foundation Models for Particle Accelerators

Frontier Language Models Struggle to Copy: Text Can Be Better Viewed in 2D

Generating the Unheard: Phylogeny-Guided Latent Generation for Ancestral Sound Reconstruction

Generative Control as Optimization: Time Unconditional Flow Matching for Adaptive and Robust Robotic Control

GeoWind2Plan: Mission-Time 3D Urban Wind Prediction for Energy-Efficient UAV Planning

Gumbo: Gumbel Optimized High-Temperature Speculative Sampling

HEED: Density-Weighted Residual Alignment for Hybrid Vision-Language Model Distillation

Heterogeneous Parallelism for Multimodal Large Language Model Training

High Entropy Regularization Leads to Symmetry Equivariant Policies in Dec-POMDPs

HiPhy: Hierarchical Alignment for Physically-Plausible Multi-Principle Video Generation

Hydra: Towards Transferable Multi-Task Learning on Temporal Graphs

Imitation from Observations with Trajectory-Level Generative Embeddings

Impacts of Aggregation on Model Diversity and Consumer Utility

Improving Context-Shift Robustness of Convolutional Networks via Context-Regularized Cross-Entropy

Inertia-1: An Open Exploration of Wearable Motion Foundation Models

Information-Directed Offline-to-Online Reinforcement Learning

Instructing LLMs to Negotiate using Reinforcement Learning with Verifiable Rewards

Invariant Features in Language Models: Geometric Characterization and Model Attribution

ITPEval: Benchmarking Formal Translation Across Interactive Theorem Provers

Joint Consistency: A Unified Test-Time Aggregation Framework via Energy Minimization

LaCache: Robust Semantic Caching for LLM Serving

LACE: Lattice Attention for Cross-thread Exploration

Lang-SVG: Hierarchical Image Vectorization with Language Priors

Language-Based Agent Control

Language-Conditioned World Modeling for Visual Navigation

Language Denoising Objectives Extend the Value of Limited Data

Language Model Goal Selection Differs from Humans' in a Self-Directed Learning Task

Learning-Augmented Online Scheduling with Parsimonious Preemption

Learning Chance-Constrained MDPs with Bellman Distributional Certificates

Learning Large-Scale Competitive Team Behaviors with Mean-Field Interactions

Learning Orthonormal Bases for Function Spaces

Learning Polyhedral Conformal Sets for Robust Optimization

Learning Theory of Transformers: Local-to-Global Approximation via Softmax Partition of Unity

Learning the Signature of Memorization in Autoregressive Language Models

Learning to Persuade a Biased Receiver

Learning to Persuade Exposes How Easily LLMs Abandon Correct Beliefs

Learning to Price with Persuasion

Learning Visual Feature-Based World Models via Residual Latent Action

LEIA: Learned Environment for Interactive Architected Materials

L-Flow: Longitudinal Flow Matching for Progression-Aware Speech Biomarker Modeling

LIBERO-PeRM: Benchmarking Personalized Robotic Manipulation

LLM-Based Multi-Agent Blackboard System for Information Discovery in Data Science

MAdam: Metric-Aware Multi-Objective Adam

Magnetic Resonance Unpaired Image Translation with Pseudometric Schrödinger Bridges

Manifold Sampling via Entropy Maximization

MaRiO: Multi-agent Collaborative Reasoning via Shared Observations in MLLMs

MarkTune: Improving the Quality-Detectability Trade-off in Model-Embedded LLM Watermarking

MaterialsSaddles: 34 Million Transition States and a Flow-Matching Saddle-Point Predictor for Materials

MC$^2$Mark: Distortion-Free Multi-Bit Watermarking for Long Messages

Measuring Safety Alignment Effects in Autonomous Security Agents

MedExAgent: Training LLM Agents to Ask, Examine, and Diagnose in Noisy Clinical Environments

Memory Retrieval for Changing Preferences

Mitigating Factual Hallucination in Large Reasoning Models via Mixed-Mode Advantage Regularization

Mitigating Object Hallucination in Large Vision-Language Models via False Discovery Controlled Visual Data Splitting

Mixture-of-Chains: Learning Causal Graphs from Human Knowledge

Model-Based Online Decision Making via Generative Trajectory Planning

MolmoMotion: Forecasting Point Trajectories in 3D with Language Instruction

MoSE3: Learning World-Space SE(3) at Every Pixel

Natural Policy Gradient as Doubly Smoothed Policy Iteration: A Bellman-Operator Framework

Near-Optimal Sample Complexity of Robust Reinforcement Learning with KL Uncertainty Set

Neglected Free Lunch from Post-training: Progress Advantage for LLM Agents

Neural Expansion: A Unified Mechanism for How Deep Neural Network Generalize

Neural Field Thermal Tomography: A Differentiable Physics Framework for Non-Destructive Evaluation

Neural Harmonic Measure Operator

Neural Modal Decomposition: Architectural Priors from Observables

NINJA: A Navigator–Inspector Joint Architecture for Context-Efficient Issue Localization

Normalizing Flows are Capable Trajectory Planners

NPUsper: Eliminating Redundant Computation for Real-Time Whisper on Mobile NPUs

nuReasoning: A Reasoning-Centric Dataset and Benchmark for Long-Tail Autonomous Driving

OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries

Offline-Online Reinforcement Learning for Linear Mixture MDPs

Online Change-point Detection using Foundation Probabilistic Forecasting Models

Online Imitation Learning for Stabilizing Vlasov--Poisson Plasmas

Online Learning in Stabilized Linear Dynamical Games with Adversarial Disturbances

Online Learning with Certified Unlearning over Graphs

On the Convergence Analysis of Muon

On the Design Space of Discrete Diffusion Online Adaptation for Molecular Optimization

On the Meta-Design of Allocation Problems

On Time, Within Budget: Constraint-Driven Online Resource Allocation for Agentic Workflows

Optimal Algorithms for Fixed-Graph Multi-Attribution Privacy

Optimal Real-Data Allocation for Synthetic-Data-Augmented Inference

Optimization Dynamics Imprint Semantic Specificity in Contrastive Embedding Norms

Optimizing Computational-Statistical Runtime for Wasserstein Distance Estimation

OptiWorld: Optimal Control for Video World Generation under Physical Constraints

Ordinary Least Squares as an Attention Mechanism

PACE: Two-Timescale Self-Evolution for Small Language Model Agents

PACEvolve++: Improving Test-time Learning for Evolutionary Search Agents

Paint Anything: Toward Any-Color Controllable Image Generation and Editing

PBT-Bench: Benchmarking AI Agents on Property-Based Testing

Perfect Parallelization in Mini-Batch SGD with Classical Momentum Acceleration

PhyMotion: Structured 3D Motion Reward for Physics-Grounded Human Video Generation

PhysicianBench: Evaluating LLM Agents in Real-World EHR Environments

PM-LoRA: Scalable Continual Learning via Progressive Merging of Low Rank Adapters

Pointwise Lipschitz Continuous Graph Algorithms

Polynomial-Time Robust Multiclass Linear Classification under Gaussian Marginals

PORT: Preference Optimization via Robust Token-Level Reweighting

Position: We Need Greater Transparency to Maintain Research Pipeline Reliability Despite GenAI

Post-hoc Selective Classification for Reliable Synthetic Image Detection

PRISM-Bench: A Benchmark of Puzzle-Based Visual Tasks with CoT Error Detection

ProcObject-10K: Benchmarking Object-Centric Procedural Understanding in Instructional Videos

Proper Agnostic Learning of Functions of Halfspaces

Q-MMR: Off-Policy Evaluation via Recursive Reweighting and Moment Matching

RAGBoost: Robust Tabular Learning via Retrieval-Augmented and Ancillary-Guided Gradient Boosting

Rank-Transformed Dissimilarity Profiles for High-Dimensional Classification

Reaching a Consensus in Predictive Loops

Reasoning Pathologies in Large Language Models: A Diagnostic Perspective

Recursive Language Models

Relational Feature Distillation for Lightweight 3D Point Cloud Segmentation

Revisiting Diffusion Model Predictions Through Dimensionality

Riemannian Ordinary Least Squares

RnR: a meta-solver for causal discovery in undersampled time series data

Robust Inference-Time Steering of Protein Diffusion Models via Embedding Optimization

Robust Instruction Compliance in Cooperative Multi-Agent Reinforcement Learning

Robust Policy Optimization to Prevent Catastrophic Forgetting

SALART-VQA: Diagnosing Whether VLMs Understand Salient Artifacts in Generated Images

Scaling Multi-Teacher Distillation for Digital Pathology

Self-Tuning Graph Filters via State-Dependent Operator Composition

Semantic Optimal Transport for Sparse Autoencoder Feature Matching and Circuit Compression

Sequential Probabilistic Uncertainty Estimation for Parallel Multi-Agent Reasoning Systems

Severity-Controlled Prediction Sets for Medication Recommendation

Sharp Convergence and Sample Complexity of Policy Mirror Descent for Average-Reward MDPs

Shattered Compositionality: Counterintuitive Learning Dynamics of Transformers for Arithmetic

SimSD: Simple Speculative Decoding in Diffusion Language Models

Sinkhorn Based Associative Memory Retrieval Using Spherical Hellinger Kantorovich Dynamics

SMASH: Probing Speech Recognition Robustness via Semantically Targeted Bit Flips

SMILE: Bridging Continuous Optimization and Discrete Symbolic Recovery

Solving Max-Cut to Global Optimality via Feasibility-Preserving Graph Neural Networks

SourceBench: Can AI Answers Reference Quality Web Sources?

Sparse Multimodal Switching State-Space Models for Regime-Dependent Neural Connectivity under Partial Observations

Sparsifying Correlation Clustering: Edge Coresets, Triangle Witnesses, and Observation Lower Bounds

Spectral Graph Sparsification Preserves Representation Geometry in Graph Neural Networks

Spectral Measures of Mamba Conductances Predict and Shape Effective Receptive Fields

Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling

Sticky Jump Diffusions: A Unifying Framework for Discrete, Continuous, and Hybrid Diffusion

Stochastic Interpolants via Conditional Dependent Coupling

StreamGaze: Gaze-Guided Temporal Reasoning and Proactive Understanding in Streaming Videos

Structured Sparse Memory for Recurrent Reasoning

SURF: Steering the Scalarization Weight to Uniformly Traverse the Pareto Front

Swift Sampling: Selecting Temporal Surprises via Taylor Series

Systematic Scaling Analysis of Jailbreak Attacks in Large Language Models

TACT-KV: Tri-Axis Cosine Transform for Compressing Volumetric KV Caches in Medical VLMs

Targeted Review for AI-Assisted Biodiversity Surveys: Active Continuous-Score Occupancy Modeling

Task Vector Geometry Underlies Dual Modes of Task Inference in Transformers

TaxaAdapter: Scaling Fine-grained Species Image Generation To the Tree of Life

TDBench: Benchmarking Vision Language Models on Top-Down Image Understanding

TENG-BC: Time-Evolving Natural Gradient for High-Accuracy Neural PDE Solvers with General Boundary Conditions

Test-Time Gradient Guidance of Flow Policies in Reinforcement Learning

The Path Not Taken: RLVR Learns Off the Principals

The Price of Locality: Why Forward-Forward Underperforms Backpropagation?

The two clocks and the innovation window: When and how generative models learn rules

The Unembedding Bottleneck: A Mechanistic Analysis of Single Digit Counting in LLMs

The Z-Gromov-Wasserstein Distance

ThinkJEPA: Empowering Latent World Models with Large Vision-Language Reasoning Model

TimeWarp: Evaluating Web Agents by Revisiting the Past

Toward a Unified Statistical Theory of Unsupervised Pretraining and Supervised Neural Knowledge Graph Learning

Towards Understanding Momentum Acceleration in River-Valley Loss Landscape

Towards Visual Query Segmentation in the Wild

Tracing Moral Foundations in Large Language Models

Training with (Swap) Regret Loss in a Single-Layer Self-Attention Model: A Case Study on the Probability Simplex

Transferable, Time-Parallel Graph Dynamics via Space-Time Factorization

Transformers Converge to Invariant Algorithmic Cores

Transformers Provably Implement In-Context Reinforcement Learning with Policy Improvement

Tree Search With Predictions

TRIAGE: Evaluating Prospective Metacognitive Control in LLMs under Resource Constraints

TriSearch: Learning to Optimize Triangulations via Bistellar Flips

Truncated Riemannian (1+1)-ES for Black-Box Optimization with Intrinsic Dimension Guarantees

Understanding Circulant Permutation to Extend the CMinHash Estimator

Understanding Private Evolution as Learning-Augmented Clustering

Uniform-in-Time Weak Propagation of Chaos in Shallow Neural Networks

Unsupervised Physics Informed Decomposition of Incomplete Time-Resolved Spectroscopy

VeriWorld: A Verifiable Visual SWE-Bench for Spatial Reasoning in 3D Environments

VeruSAGE-Bench: A Benchmark Suite for Rust System Verification

Vibe-Spike: An Energy-Preserving EEG Foundation Model Through the Landscape of Neural Coherence

Video Forensic Self-Descriptions: Leveraging Temporally Distributed Forensic Microstructures for Zero-Shot Detection of AI-Generated Videos

VideoMLA: Low-Rank Latent KV Cache for Minute-Scale Autoregressive Video Diffusion

Video Models Can Reason with Verifiable Rewards

VLA-REPLICA: A Low-Cost, Reproducible Benchmark for Real-World Evaluation of Vision-Language-Action Models

VTV-FM: Flow Matching through Variational Terminal-Velocity Closure

WavFlow: Flowing Through Waveforms for Audio Generation

What Does an Observability Forecasting Foundation Model Know?

What Fits (Into Few Tokens) Doesn't Overfit: Compression and Generalization in ML Research Agents

What Sound Tells You About the Room

When Are Semivalue-Based Decisions Identifiable? Robust Data Selection under Utility Ambiguity

When Does Subspace Direction Matter for LoRA? Regime Analysis of the Magnitude Principle in Few-Shot Adaptation

When Everyone Can Submit: Designing Contests with Transparent Pre-Selection

When Reasoning Meets Its Laws

Which Tokens to Merge? Diffusion Dynamics for Efficient Image Generation

You Can’t Have It Both Ways: Concept Entanglement Limits Diffusion Model Unlearning

(ends 5:00 AM)

4:30 a.m.

Break:

Retail Lunch

(ends 6:30 AM)

7:30 a.m.

Atlanta Orals 3A World Models, 3D Scenes & Collaborative Agents [7:30-8:30]

Orals 7:30-8:30

[7:30] Language-Conditioned World Modeling for Visual Navigation

[7:30] SyncWorld: Visual Calibration Enables World Models as Zero-Shot Simulators

[7:30] TransmissiveGS: Residual-Guided Disentangled Gaussian Splatting for Transmissive Scene Reconstruction and Rendering

[7:30] Training Generalizable Collaborative Agents via Strategic Risk Aversion

(ends 8:30 AM)

Atlanta Orals 3B Benchmarks & Evaluation Methodology [7:30-8:30]

Orals 7:30-8:30

[7:30] MLS-Bench: A Holistic and Rigorous Assessment of AI Systems on Building Better AI

[7:30] HearSayBench: Can LLMs Navigate from Abstract Human Rights to Lived Lives?

[7:30] MOVEBENCH: A Benchmark for Global-Scale Wildlife Movement Forecasting

[7:30] DD-Ranking: Rethinking the Evaluation of Dataset Distillation

(ends 8:30 AM)

8:30 a.m.

Break:

Coffee & Snacks

(ends 9:00 AM)

Atlanta Poster Session 6 [8:30-11:30]

Posters 8:30-11:30

ABC-Align: Prediction-Powered Alignment with Adaptive Bias Control

ABC: Any-Subset Autoregression via Non-Markovian Diffusion Bridges in Continuous Time and Space

Adapting in the Dark: Efficient and Stable Test-Time Adaptation for Black-Box Models

A Deep Learning Framework for Scalar-on-Function Models

AgentAbstain: Do LLM Agents Know When Not to Act?

Agent-Native Research Artifacts

Agents' Last Exam

AI Agents Push Humans Out of the Loop

AI GAMESTORE: Scalable, Open-Ended Evaluation of Machine General Intelligence with Human Games

Align-RAG: Alignment Is All You Need for TSFM In-Context Learning

A Multi-Fidelity Control Variate Approach for Policy Gradient Estimation

A Multimodal Benchmark for Evaluating Cause-of-Death Inference Using Child Health and Mortality Data

ANCRe: Adaptive Neural Connection Reassignment for Efficient Depth Scaling

A Principled Self-Referenced Early Stopping Approach for Deep Image Prior

A Process-Level Evaluation of LLM Discovery Agents

Arena-T2I Hard: Benchmarking and Improving Faithfulness with Dependency-Aware Checklist Rewards

A Scalable Measure of Loss Landscape Curvature for Analyzing the Training Dynamics of LLMs

A second order regret bound for NormalHedge

AstraFlow: Dataflow-Oriented Reinforcement Learning for Agentic LLMs

AsymHP: Load-Balanced Sparse Attention for Video Diffusion Transformers

Asymmetric Phase Coding Audio Watermarking

A Topological Encoder Decoder Framework for Temporal Graph Learning

Auditing Cross-Lingual Fairness in Language Model Watermarking

Auditor-Assisted Summary-Channel Verification for Hosted LLM Identity Substitution

Automated Hypothesis Discovery for Characterizing Annotation Disagreement

Automatically Refining Coding Rules for AI Coding Agents

AVI-HT: Adaptive Vision-IMU Fusion for 3D Hand Tracking

AVSD: Adaptive-View Self-Distillation by Balancing Consensus and Teacher-Specific Privileged Signals

B2P-Corr: Batch-to-Population Gradient Estimators for Non-Decomposable Correlation Losses

Base Items Overfit, New Items Underfit: Hidden Cost of Joint Training in Incremental Adaptation

Bayesian Preference Learning for Test-Time Steerable Reward Models

Bayes-pFCL:Bayesian Personalized Federated Continual Learning

Benchmarking Fine-Grained Spatio-Temporal Awareness in Embodied Brain Models

Benchmarking Risk Attitudes of LLMs

Beyond IPS: Reliable Counterfactual Evaluation in Multi-Stage Ad Systems without Logged Propensities

Beyond Risky Activities: Bridging the Supervision Gap for Situational Risk Reasoning

Bias, Measurement Error, and Double-Dipping: When Can GNN Convolutions Help Brain Connectome Prediction?

BOOST: Power-Optimal Strong-FWER Testing for Block-Structured Multiplicity

Borda-Based Fair Multi-User Dueling Bandit in Tabular and Generalized Linear Settings

BSTabDiff: Block-Subunit Diffusion Priors for High-Dimensional Tabular Data Generation

Calibrating LLMs with Semantic-level Reward

CASL: Concept-Aligned Sparse Latents for Interpreting Diffusion Models

CAST: Causal Anchored Simplex Transport for Distribution-Valued Time Series

Cat-DPO: Category-Adaptive Safety Alignment

Class-Mixed Diffusion Augmentation for Shortcut-Breaking in Continual Learning

Combating Catastrophic Forgetting in Continual Domain Adaptation via Knowledge Recasting

Composing Diffusion Priors with Explicit Physical Context via Generative Gibbs Sampling

Concurrence of Symmetry Breaking and Nonlocality Phase Transitions in Diffusion Models

ConQuR: Corner Aligned Activation Quantization via Optimized Rotations for LLMs

Conservation Laws for Diffusion Models

Continuity Laws for Sequential Models

ContractBench: Can LLM Agents Preserve Observation Contracts?

Contrast encodes inductive bias: separating slow noise from dynamics in predictive representation learning

Cornerstones or Stumbling Blocks? Deciphering the Rock Tokens in On-Policy Distillation

Corrected Integrated Laplace Approximation for Bayesian Inference in Latent Gaussian Models

Correlating Cross-Iteration Noise for DP-SGD using Model Curvature

CoScan: Multi-Scale Content-Adaptive Space-Filling Scans for Causal State-Space Image Restoration

Cross-Attentive Bayesian Low-Rank Adaptation for Multimodal Uncertainty Estimation

Cross-Dialect Generalization Without Retraining: Benchmarks and Evaluation of Schema-Derived Constrained Decoding for MLIR

CultureRed: Benchmarking Culture-Specific AI Safety Based on Global Statutes

DBPS: Doob-Bridge Posterior Sampler with Balanced Endpoint-Population Control for Unpaired Neurodegenerative Pathology Transport

DeepArrhythmia: Segment-Contextualized ECG Arrhythmia Classification via Selective Evidence Acquisition

Deep-Koopman-KANDy: Dictionary Discovery for Deep-Koopman Operators with Kolmogorov-Arnold Networks for Dynamics

Default Feature Representations of the Cognitive Map

Denoising Distances in Metric Measure Spaces

Deployment-Memory LLM Test-Time Training Should Require Behavioral Evidence Beyond Perplexity

Developmental Visual Experience Scaffolds Grounded Concept Acquisition in Vision-Language Models

Diffusion-DRF: Free, Rich, and Differentiable Reward for Video Diffusion Fine-Tuning

Direct Acceleration of Stochastic Root-Finding Without Variance Reduction and Regularization

Discovering Symbolic Differential Equations with Symmetry Invariants

Disentangling Where and When: Factored Spatio-Temporal Explanations for Video Action Recognition

Diversity Maximization: Algorithms for Distant $k$-Subsets

DnNP: Denoising Input Uncertainty in Neural Processes

Driving Video Retrieval for Complex Queries with Structured Grounding

DRScaffold: Boosting Dense-Scene Reasoning in Lightweight Vision Language Models

Dual Dimensionality for Local and Global Attention

Dynamic Treatment on Networks

E4GEN: Event-level Explainable Extreme-Enhanced Time-series Generation

Escaping the Cognitive Well: Efficient Competition Math with Off-the-Shelf Models

Evidential Semantic Uncertainty Decomposition for Large Language Models

Evolutionary Feature Engineering for Structured Data

Evolving Agent Teams

Extending 3D Reconstruction Models to Any Camera

Factorized Spectral Representations for Reinforcement Learning

Feature Learning Dynamics in Infinite-Depth Neural Networks

Feature learning in high-dimensions under structured covariance: Scaling laws in quadratic networks

FedIndex: Federated Domain Adaptation with Continuous Domain Indices

FindStatBench: Evaluating Large Language Models on Combinatorial Code Synthesis

Finite-Memory Control of POMDPs: Fundamental Limits and Efficient Design

Fisher-R1: Training LLM Agents for Reliable Hypothesis Testing

FlashBoB: I/O-Efficient Exact Backward-over-Backward for Softmax Attention

FluxLite: Inference-Time Proposal Control for Discrete Diffusion Models

Forced Deferral: Manipulating Routing Decisions in Multimodal LLM Cascades

Forward Shapley Scoring for Non-Myopic Active Feature Acquisition

Found in Conversation: LLMs Teach Themselves to Close the Multi-Turn Gap

From Heartbeat to Cardiochoreography: A Mechanics Foundation Model for Individualized 4D Cardiac Motion Generation Conditioned on Electrophysiology

From Preference Data to Personalization: Tracing Sycophancy in Large Language Models

From Talking Words to Sharing Thoughts: Scalable Multi-LLM Aggregation via Structured Message Passing

From Weeks to Hours: Fast and Principled SFT Curation for LLM

Generative Cross-Entropy: A Strictly Proper Loss for Data-Efficient Classification

GenScale: A Benchmark for Relative Object Scale in Image Generation and Editing

Geometric Signatures of Reasoning: A Spectral Perspective on Task Hardness

GeoSPRINT: Geometric Redundancy-Aware Step Pruning for Inference in Diffusion Trajectories

GLOBE: Accurate Surrogates for Boundary-Driven PDEs via Domain-Inspired Architectures and Equivariance

GOLD PANNING: Strategic Context Shuffling for Needle-in-Haystack Reasoning

GPT-Image-Edit-1M: An Auditable Million-Scale Dataset for Instruction-Guided Image Editing

Gradient descent inference in empirical risk minimization

GRAM: Group-wise Rank-Aware Modal Merging via Subspace Alignment

Grounded-Exo2Ego: Structured Semantic Grounding for Robust Exocentric-to-Egocentric Video Generation

Grounding Multimodal Reasoning with Evidence-Ablated Negatives

Guaranteed Noisy CP Tensor Recovery via Riemannian Optimization on the Segre Manifold

Guarding the Life Code: Preserving Membership Privacy in Genomic Foundation Models

Heavy-Tailed Flow Matching via Random Clocks

High Probability Risk Control for Online Policy Learning

How to Have a Sensitive Debate

HybridCache: Enhancing Prefix Caching for Linear–Softmax Language Models

HyCO: A Hybrid Neural Solver for Combinatorial Optimization

Improving Conditional Modeling via Inter-Class Likelihood-Ratio Maximization and Unifying Classifier-Free Guidance with Alignment Objectives

Improving Function Space Flow Matching with Kernel Optimal Transport

In-Context Optimization for Retrieval-Augmented Generation: A Gradient-Descent Perspective

InstructVVT: Instruction-Driven Video Virtual Try-On without Auxiliary Spatial Priors

Interpreting Neural Combinatorial Optimization via Evolving Programmatic Bottlenecks

Invariance and Body-Order Compose Additively: Minimax Rates on $\mathrm{SO}(3)^n$

JobBench: Aligning Agent Work With Human Will

JudgmentBench: Comparing Rubric and Preference Evaluation for Quality Assessment

Latent Barrier Steering: Hierarchical Safety for Generative Planning

LEAD: Length-Efficient Adaptive and Dynamic Reasoning for Large Language Models

Lean4Agent: Formal Modeling and Verification for Agent Workflow and Trajectory

Learning-Augmented Mechanism Design for Facility Location under $L_p$-Norm Social Costs

Learning from Language Feedback via Variational Policy Distillation

Learning Interpretable Switching Dynamics in Shared Neural-Behavioral Latent Space

Learning Process Rewards via Visitation Matching for Efficient RL

Learn Locally, Recurse Globally: Neural Circuit Synthesis Beyond Training Depth

LEMON-ZEST: Evolution-Informed Tokenization for Efficient Protein Language Modeling

Likelihood-Free Generative Policy Optimization

Linear Regression Under Misalignment: Algorithms and Theoretical Results

Localizing and Repairing Sparse-Prompt Failure in SAM Decoders via Box-to-Point Counterfactual

Local linear convergence of gradient methods for overparameterized Gaussian mixtures

LogT: Logically Think with Images for Visual Search

Loop alignment: Self-organized Weight Transpose in Predictive Coding through Independent Hebbian Plasticity.

MAGNET: Manifold-Aware Graph Diffusion Network for Connectome Generation

Markovian Experimental Design under Concept Drift

MARS: Harmonizing Multimodal Convergence via Adaptive Rank Search

MathlibPR: Pull Request Merge-Outcome Benchmark for Formal Mathematical Libraries

MCM-DM: Towards Better Spatio-Temporal Event Representation Learning via Discrete Morse Theory

MCP-Atlas: A Large-Scale Benchmark for Tool-Use Competency with Real MCP Servers

MemLeak: Diagnosing Information Leaks in Multimodal Agent Memory

Missing data and cluster graphs: cluster-level missingness vs variable-level missingness

Mixture of Layers: Dynamic Layer Routing for Visual Reasoning

MLLMs Fail to Refuse when Using Tools Agentically

MoMHa: Multi-Objective Optimization of LLM Harnesses over Accuracy, Safety, and Tokens

Monoculture Robust Learning

MOSAIC-CONUS: A Multimodal, Multi-Temporally Paired dataset for Earth Sciences

MOSAIC: Module Discovery via Sparse Additive Identifiable Causal Learning for Scientific Time Series

Motion-o: Trajectory-Grounded Video Reasoning

MPQ: A Message-Passing View of Post-Training Quantization

Multidimensional Observer Model and Perceptual Dimensions of Human Image Quality Assessment

NitroBox: Lightning-Fast Sandbox for Large-Scale RL Training

Nonconvex Decentralized Stochastic Bilevel Optimization under Heavy-Tailed Noise

Non-Linear Pricing Restores Tractability for a Data Seller

Normalizing Trajectory Models

OASES: Outcome-Aligned Search-Evaluation Co-Training for Agentic Search

OATS: Online Data Augmentation for Time Series Foundation Models

OmniMechanism Design for Human-AI Collaboration with Impressionable Minds

OmniMem: Scalable and Adaptive Memory Retrieval for Long Video Generation

OmniShotCut: Holistic Relational Shot Boundary Detection with Shot-Query Transformer

One Model, Two Roles: Emergent Specialization in a Shared Recurrent Transformer

Online Learning on Hidden-Convex Losses via Algorithmic Equivalence: Optimal Regret, Geometric Barrier, and Bandit Feedback

On the Fragility of Latent Knowledge: Layer-wise Influence under Unlearning in Large Language Model

On the Geometry and Latent-Space Composition of Hypernetwork-Generated LoRAs

On the Invariance and Generality of Neural Scaling Laws

Opening the Black Box of Classifier-Free Guidance via Information Bottleneck

Optimal Learning-Augmented Algorithm for Online Bidding

Optimal sequential tests yield log-optimal e-processes

Oracle Supervision Transfers for Hyperparameter Prediction in Model-Based Image Denoising

PAAC: Privacy-Aware Agentic Device-Cloud Collaboration

Panoptic Scene Program Diffusion Transformer

Paradoxical noise preference in RNNs

PatternBloom: Empowering Agentic RAG with Externalized RL-Distilled Graph Patterns

Personal Visual Memory from Explicit and Implicit Evidence

PerturbReason: A Knowledge-Grounded Benchmark and Framework for Cell-State–Conditioned Mechanistic Reasoning of Perturbation Effects

PhyMetric: Diagnosing Physical Plausibility in Text-to-Video Generation via Scene-Level QA

Physical AI Smart Spaces: A Large-Scale Benchmark for Multi-Camera 3D Perception in Smart Spaces

PM1: A Multimodal Foundation Model for Genomes, Phenotypes, and Images at Biobank Scale

PocketVE: Stable and Controllable Structure-Based Drug Design with Variance-Exploding Diffusion

Policy Gradient Algorithms in Average-Reward Multichain MDPs

Policy Optimization in Hybrid Discrete-Continuous Action Spaces via Mixed Gradients

PolyMind: Exploring Width Scaling for Reflective Reasoning in Language Agents

PortPy: A Benchmark for AI and Optimization in Cancer Radiotherapy Planning

Precise but Uncoupled: Reviewer Precision Does Not Guarantee Critique Uptake in Multi-Agent Math Reasoning

Prediction-only distillation with optimal mixing in ridge-regularized linear and logistic regression

Predictive 4D Generation with Latent State Machines

Progressive Risk Estimation for Accident Anticipation

Projected Neural Additive Models as Universal Approximators

Prototypes of the Mind: A Unified Framework for Probing the Visual Brain

Provable Quantization with Randomized Hadamard Transform

Provable Test-Time Scaling for Beam Search in LLM Reasoning

Prudent-Banker: No Extra Fees for Baseline Safety in Adversarial Bandits With and Without Delays

QueryStop: Dynamic Stop Signals for Efficient Streaming Inference

RaZeR: Pushing the Limits of NVFP4 Quantization with Redundant Zero Remapping

ReaLM: A Unified Red-Teaming Benchmark for Physical-World VLMs

Reasoning with Sampling: Cutting at Decision Points

Representing Part-Whole Hierarchy with Nested Neuronal Coherence

Rethinking Language Model Scaling under Transferable Hypersphere Optimization

Rethinking Visual Attribution for Chest X-ray Reasoning in Large Vision Language Models

ReToken: One Token to Improve Vision–Language Models for Visual Retrieval

Reward Bias Substitution: Single-Axis Bias Mitigations Redirect Optimization Pressure

Riemannian Lyapunov Framework: Optimization as Closed-Loop Control on Riemannian Manifolds

Risk-Calibrated Context Selection for Healthcare Multi-Agent Handoff

RoMo Hands: A Large Scale Richly Organized Text to Hand Motion Dataset

RoPEMover: Depth-Aware Object Relocation via Positional Embeddings

RoSeViT: Role-Separated Vision Transformers for ARC Visual Reasoning

Routing as a Singular Reparameterization: A Closed-Form Pushforward Prior and Exact Bayesian Complexity in a Minimal Proxy

RTEB: An Overfitting-Resistant Benchmark for Embedding Model Evaluation

RuleSmith: Multi-Agent LLMs for Automated Game Balancing

Runtime Monitoring of Perception-Based Autonomous Systems via Embedding Temporal Logic

Runtime Verification of Multiple Natural Language Criteria for Agent Governance

Saliency-Aware Multi-Route Thinking: Grounding and Reasoning on Vision-Language Agents

SAMPPO: Structure-Aware Mirror Proximal Policy Optimization

SapiensID 2.0: Aligning Human Recognition Foundation Models with Human Perception

SARA: Step-Adaptive Rank Adjustment for Diffusion Inference Acceleration

SASA: Subspace-Aware Sparse Autoencoders for Effective Mechanistic Interpretability

ScaleBITS: Scalable Bitwidth Search for Hardware-Aligned Mixed-Precision LLMs

Scaling Whole-Body Loco-Manipulation through Compositional Data Synthesis

SCOUT: Planning under Occlusion via Object-Centric World Model Rollouts

Self-Consuming Generative Models with Co-Evolving Human Preferences

Self-Supervised Doppler-Guided RF Odometry

Self-Supervised Keyframe Discovery for Horizon-Invariant Behavior Cloning

Semantic Search over 9 Million Mathematical Theorems

Shared Truth: Emergent Truth Properties in Large Language Models via Heterogeneous Injection-based Transfer

She Performs Your Voice: A Unified Speech and Dance Motion Model

SILSA: Sliding-Window Slice Latents for Topology-Preserving High-Resolution 3D Generation

Simple KNN-Based Outlier Detection Achieves Robust Clustering

Skill-Coupled Policy Optimization with Calibrated Group-Wise Advantage Estimation

SLIC: Reinforcement Fine-Tuning Small LMs for Multi-Turn Analog Circuit Optimization

SOL-ExecBench: Speed-of-Light Benchmarking for Real-World GPU Kernels Against Hardware Limits

Space-Optimal Streaming Algorithms via Efficient Encodings

SplitZip: Ultra Fast Lossless KV Compression for Disaggregated LLM Serving

Stability in Multi-Step Reasoning via Jacobian-based Error Accumulation Analysis

STARFlow2: Bridging Language Models and Normalizing Flows for Unified Multimodal Generation

State-of-art minibatches via novel DPP kernels: discretization, wavelets, and rough objectives

Stochastic Dynamic Barrier Perturbed Gradient Methods for Nonconvex Simple Bilevel Optimization

Streaming Interventions: Can Video LLMs Correct Mistakes as They Occur?

Structured Human-Like Agentic Flow for RTL Design

Submodular Clustering beyond $1-1/e$

Taming the Curses of Multiagency in Robust Markov Games with Large State Space through Linear Function Approximation

Temporal Gradient Inversion for Private Trajectory Reconstruction in Embodied Reinforcement Learning

Test-Time Speculation

The Illusion of Forgetting: Rank Leakage in Knowledge Editing and Its Mitigation

The Query Complexity of Local Search in Rounds on General Graphs

The Sharp Directions Are Against You: Curvature Analysis of Activation Steering

TIDE: Every Layer Knows the Token Beneath the Context

Token Time Continuous Diffusion for Language Modeling

Total Variation Rates for Riemannian Flow Matching

Towards Settling the Complexity of Non-Euclidean Parallel Convex Optimization

TRACER: Token ReAssignment for Concept ERasure in Generative Recommendation

Training Generalizable Collaborative Agents via Strategic Risk Aversion

Training Transformers for KV-Cache Compressibility

Transcoda: End-to-End Zero-Shot Optical Music Recognition via Data-Centric Synthetic Training

TraXion: Rethinking Pre-training Frameworks for Mobility and Beyond

Trellis4D: Complex 4D Mesh Generation

Truthful Calibration Errors for Multi-Class Prediction

Unified High-Probability Analysis of Stochastic Variance-Reduced Estimation

Unifying Reasoning and Planning through Energy Minimization

VASR: Variance-Aware Systematic Resampling for Diffusion Models

Verifying Neural Networks with Reinforcement Learning

Viverra: Text-to-Code with Guarantees

VLS: A Vision-Language-Shape Model for Open-Vocabulary Partonomic 3D Reconstruction

VLSplat: Vision-Language Guided Object-Centric 3D Gaussian Splatting via Scene Graph

Watermarking as a Learned Intrinsic Property of Diffusion Models

Wavefunction Flows: Efficient Quantum Simulation of Continuous Flow Models

What Should Embeddings Embed? Autoregressive Models Represent Latent Generating Distributions

When Helpfulness Becomes Sycophancy: Sycophancy is a Boundary Failure Between Social Alignment and Epistemic Integrity in Large Language Models

When Latents Forget Pixels: Restoring Fidelity in Diffusion Transformer Super-Resolution

When Safety Becomes An Outlier: Understanding the Retention of LLM Safety Behaviors

When Uncertainty Is the Target: Adversarial Attacks on Uncertainty-Aware Predictors

Whole-Body Compliant Control via Learned Force-Regulation Modules

Who, Where, and What? Forensic Localization in LLM-Based Multi-Agent Systems

Witness Overlap: Directional Provenance Inside Open-Weight Model Families

Youdunit: Single-Call Counterfactual Necessity in Multi-Agent LLM Systems

(ends 11:30 AM)

11 a.m.

SUN 13 DEC

1 a.m.

Break:

Coffee & Snacks

(ends 1:30 AM)

3:30 a.m.

Break:

Retail Lunch

(ends 5:00 AM)

7:30 a.m.

Break:

Coffee & Snacks

(ends 8:00 AM)

9 a.m.

Reception:

Workshop Reception

(ends 11:00 AM)

MON 14 DEC

1 a.m.

Break:

Coffee & Snacks

(ends 1:30 AM)

4 a.m.

Break:

Retail Lunch

(ends 5:30 AM)

7:30 a.m.

Break:

Coffee & Snacks

(ends 8:00 AM)