Fri 5:30 a.m. - 5:50 a.m.
|
Breakfast
(
Breakfast
)
>
|
馃敆
|
Fri 5:50 a.m. - 6:00 a.m.
|
Opening Remarks
(
Opening
)
>
|
馃敆
|
Fri 6:00 a.m. - 6:30 a.m.
|
Fine-grained Interactive Vision Language Pre-training
(
KeyNote Talk
)
>
SlidesLive Video
|
Lu Hou 路 Lu Hou
馃敆
|
Fri 6:30 a.m. - 7:05 a.m.
|
鈥婨fficiency Tradeoffs in the Design of Neural Search Systems
(
KeyNote Talk
)
>
SlidesLive Video
|
Jimmy Lin
馃敆
|
Fri 7:05 a.m. - 7:35 a.m.
|
Last Advances in End-to-End Speech Recognition
(
KeyNote Talk
)
>
|
Tara Sainath
馃敆
|
Fri 7:35 a.m. - 7:45 a.m.
|
Collective Knowledge Graph Completion with Mutual Knowledge Distillation
(
Spotlight
)
>
SlidesLive Video
|
Weihang Zhang 路 Ovidiu Serban 路 Jiahao Sun 路 Yike Guo
馃敆
|
Fri 7:45 a.m. - 7:56 a.m.
|
Attribute Controlled Dialogue Prompting
(
Spotlight
)
>
SlidesLive Video
|
Runcheng Liu 路 Ahmad Rashid 路 Ivan Kobyzev 路 Mehdi Rezaghoizadeh 路 Pascal Poupart
馃敆
|
Fri 7:56 a.m. - 8:05 a.m.
|
Fast DistilBERT on CPUs
(
Spotlight
)
>
SlidesLive Video
|
Haihao Shen 路 Ofir Zafrir 路 Bo Dong 路 Hengyu Meng 路 Xinyu Ye 路 Zhe Wang 路 Yi Ding 路 Hanwen Chang 路 Guy Boudoukh 路 Moshe Wasserblat
馃敆
|
Fri 8:00 a.m. - 8:30 a.m.
|
Morning Break and Poster Session 1
(
Break and Poster Session
)
>
|
馃敆
|
Fri 8:30 a.m. - 9:05 a.m.
|
SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models
(
KeyNote Talk
)
>
SlidesLive Video
|
Song Han
馃敆
|
Fri 9:05 a.m. - 9:35 a.m.
|
Building Language Models Based on Retrieval
(
KeyNote Talk
)
>
SlidesLive Video
|
Danqi Chen
馃敆
|
Fri 9:35 a.m. - 10:05 a.m.
|
Colossal-AI: A Unified Deep Learning System For Large-Scale Parallel Training
(
KeyNote Talk
)
>
SlidesLive Video
|
Yang You
馃敆
|
Fri 10:05 a.m. - 10:15 a.m.
|
Efficient Few-Shot Learning Without Prompts
(
Spotlight
)
>
SlidesLive Video
|
Oren Pereg 路 Daniel Korat 路 Moshe Wasserblat 路 Lewis Tunstall 路 Unso Eun Seo Jo 路 Luke Bates 路 Nils Reimers
馃敆
|
Fri 10:15 a.m. - 10:25 a.m.
|
PCFG-based Natural Language Interface Improves Generalization for Controlled Text Generation
(
Spotlight
)
>
SlidesLive Video
|
Jingyu Zhang 路 Jim Glass 路 Tianxing He
馃敆
|
Fri 10:25 a.m. - 10:35 a.m.
|
PromptDA: Label-guided Data Augmentation for Prompt-based Few Shot Learners
(
Spotlight
)
>
SlidesLive Video
|
Canyu Chen 路 Kai Shu
馃敆
|
Fri 10:30 a.m. - 11:30 a.m.
|
Lunch Break and Virtual Poster Session
link
|
馃敆
|
Fri 11:30 a.m. - 12:00 p.m.
|
Efficient Identify Event Causality with Knowledge and Analogy
(
KeyNote Talk
)
>
SlidesLive Video
|
Bang Liu
馃敆
|
Fri 12:00 p.m. - 12:50 p.m.
|
Interactive Industrial Panel
(
Discussion Panel
)
>
SlidesLive Video
|
Jiahao Sun 路 Ahmed Ibrahim 路 Marjan Ghazvininejad 路 Yu Cheng 路 Boxing Chen 路 Mohammad Norouzi 路 Rahul Gupta
馃敆
|
Fri 12:50 p.m. - 12:59 p.m.
|
Improving the Robustness of DistilHuBERT to Unseen Noisy Conditions via Data Augmentation, Curriculum Learning, and Multi-Task Enhancement
(
Spotlight
)
>
SlidesLive Video
|
Heitor Guimar茫es 路 Arthur Pimentel 路 Anderson R. Avila 路 Mehdi Rezaghoizadeh 路 Tiago H Falk
馃敆
|
Fri 12:59 p.m. - 1:05 p.m.
|
Gradient Knowledge Distillation for Pre-trained Language Models
(
Spotlight
)
>
SlidesLive Video
|
Lean Wang 路 Lei Li 路 Xu Sun
馃敆
|
Fri 1:00 p.m. - 1:30 p.m.
|
Break and Poster Session II
(
Break and Poster Session
)
>
|
馃敆
|
Fri 1:30 p.m. - 2:05 p.m.
|
Neuro-Symbolic Language Modeling with Automaton-augmented Retrieval
(
KeyNote Talk
)
>
SlidesLive Video
|
Graham Neubig
馃敆
|
Fri 2:05 p.m. - 2:35 p.m.
|
Do we still need inductive biases after Transformer language models?
(
KeyNote Talk
)
>
SlidesLive Video
|
Siva Reddy
馃敆
|
Fri 2:35 p.m. - 3:05 p.m.
|
8-bit Methods for Efficient Deep Learning
(
KeyNote Talk
)
>
SlidesLive Video
|
Tim Dettmers
馃敆
|
Fri 3:05 p.m. - 3:35 p.m.
|
Efficient Controllable Generative Models for Music and Performance Synthesis
(
KeyNote Talk
)
>
SlidesLive Video
|
Cheng-Zhi Anna Huang
馃敆
|
Fri 3:35 p.m. - 3:45 p.m.
|
Best Paper and Poster Awards
(
Closing remark
)
>
SlidesLive Video
|
馃敆
|
-
|
Parameter-Efficient Low-Resource Dialogue State Tracking by Prompt Tuning
(
Poster
)
>
SlidesLive Video
|
Mingyu Derek Ma 路 Jiun-Yu Kao 路 Shuyang Gao 路 arpit gupta 路 Di Jin 路 Tagyoung Chung 路 Nanyun Peng
馃敆
|
-
|
BERT on a Data Diet: Finding Important Examples by Gradient-Based Pruning
(
Poster
)
>
SlidesLive Video
|
Mohsen Fayyaz 路 Ehsan Aghazadeh 路 Seyed MohammadAli Modarressi 路 Mohammad Taher Pilehvar 路 Yadollah Yaghoobzadeh 路 Samira Ebrahimi Kahou
馃敆
|
-
|
Pre-Training a Graph Recurrent Network for Language Representation
(
Poster
)
>
SlidesLive Video
|
Yile Wang 路 Linyi Yang 路 Zhiyang Teng 路 Ming Zhou 路 Yue Zhang
馃敆
|
-
|
An Efficient Memory-Augmented Transformer for Knowledge-Intensive NLP Tasks
(
Poster
)
>
SlidesLive Video
|
Yuxiang Wu 路 Yu Zhao 路 Baotian Hu 路 Pasquale Minervini 路 Pontus Lars Erik Saito Stenetorp 路 Sebastian Riedel
馃敆
|
-
|
QuaLA-MiniLM: a Quantized Length Adaptive MiniLM
(
Poster
)
>
SlidesLive Video
|
Shira Guskin 路 Moshe Wasserblat 路 Haihao Shen 路 Chang Wang
馃敆
|
-
|
Towards Data Efficient And Robust Speech Representation Model Distillation
(
Poster
)
>
SlidesLive Video
|
Pheobe Sun 路 Ruibo Shi 路 Ahmad Emami 路 Sean Moran
馃敆
|
-
|
On Spectral and Temporal Feature Encoding Behaviour in Stacked Architectures
(
Poster
)
>
SlidesLive Video
|
Vaibhav Singh 路 Vinayak Abrol 路 Karan Nathwani
馃敆
|
-
|
Few-Shot Aspect Extraction using Prompt Training
(
Poster
)
>
SlidesLive Video
|
Oren Pereg 路 Daniel Korat 路 Moshe Wasserblat 路 Kfir Bar
馃敆
|
-
|
Can we get smarter than majority vote? Efficient use of individual rater鈥檚 labels for content moderation
(
Poster
)
>
|
Changho Shin 路 Alice Schoenauer-Sebag
馃敆
|
-
|
BudgetLongformer: Can we Cheaply Pretrain a SOTA Legal Language Model From Scratch?
(
Poster
)
>
SlidesLive Video
|
Joel Niklaus 路 Daniele Giofr猫
馃敆
|
-
|
Parameter-Efficient Finetuning of Transformers for Source Code
(
Poster
)
>
SlidesLive Video
|
Shamil Ayupov 路 Nadezhda Chirkova
馃敆
|
-
|
Graph Masking Pre-training for Graph-to-Text Generation
(
Poster
)
>
SlidesLive Video
|
Jiuzhou Han 路 Ehsan Shareghi
馃敆
|
-
|
The Ineffectiveness of TKGE Models in Encoding Real-World Knowledge Graphs
(
Poster
)
>
SlidesLive Video
|
Chuan Ming Ong 路 Jiahao Sun 路 Ovidiu Serban 路 Yike Guo
馃敆
|
-
|
PEST: Combining Parameter-Efficient Fine-Tuning with Self-Training and Co-Training
(
Poster
)
>
SlidesLive Video
|
Hunter Lang 路 Monica Agrawal 路 Yoon Kim 路 David Sontag
馃敆
|
-
|
ContextNER: Contextual Phrase Generation at Scale
(
Poster
)
>
SlidesLive Video
|
Himanshu Gupta 路 Shreyas Verma 路 Tarun Kumar 路 Swaroop Mishra 路 Tamanna Agrawal 路 Amogh Badugu 路 Himanshu Bhatt
馃敆
|
-
|
Efficient Speech Translation with Pre-trained models
(
Poster
)
>
SlidesLive Video
|
Zhaolin Li 路 Jan Niehues
馃敆
|
-
|
Dynamic Query Representation for Extractive Question Answering
(
Poster
)
>
SlidesLive Video
|
Urchade Zaratiana 路 Niama El Khbir 路 Dennis N煤帽ez-Fern谩ndez 路 Pierre Holat 路 Nadi Tomeh 路 Thierry Charnois
馃敆
|
-
|
Strategies for Applying Low Rank Decomposition to Transformer-Based Models
(
Poster
)
>
SlidesLive Video
|
Habib Hajimolahoseini 路 Walid Ahmed 路 Mehdi Rezaghoizadeh 路 Vahid Partovi Nia 路 Yang Liu
馃敆
|
-
|
DyLoRA: Parameter Efficient Tuning of Pre-trained Models using Dynamic Search-Free Low Rank Adaptation
(
Poster
)
>
SlidesLive Video
|
Mojtaba Valipour 路 Mehdi Rezaghoizadeh 路 Ivan Kobyzev 路 Ali Ghodsi
馃敆
|
-
|
Pyramid Dynamic Inference: Encouraging Faster Inference via Early Exit Boosting
(
Poster
)
>
SlidesLive Video
|
Ershad Banijamali 路 Pegah Kharazmi 路 Samridhi Choudhary 路 Sepehr Eghbali 路 Clement Chung
馃敆
|
-
|
An efficient RNN Language Model using activity sparsity and sparse back-propagation through time
(
Poster
)
>
SlidesLive Video
|
Mark Schoene 路 Khaleelulla Khan Nazeer 路 David Kappel 路 Christian Mayr 路 Anand Subramoney
馃敆
|
-
|
An Exploration of Methods for Zero-shot Transfer in Small Language Models
(
Poster
)
>
SlidesLive Video
|
Alon Albalak 路 Akshat Shrivastava 路 Chinnadhurai Sankar 路 Adithya Sagar 路 Mike Ross
馃敆
|
-
|
On the impact of the quality of pseudo-labels on the self-supervised speaker verification task
(
Poster
)
>
SlidesLive Video
|
Abderrahim Fathan 路 JAHANGIR ALAM 路 Woo Hyun Kang
馃敆
|
-
|
INT8 Transformers for Inference Acceleration
(
Poster
)
>
SlidesLive Video
|
Andy Rock 路 Omar Khalil 路 Ofer Shai 路 Paul Grouchy
馃敆
|
-
|
Parameter and Data Efficient Continual Pre-training for Robustness to Dialectal Variance in Arabic
(
Poster
)
>
SlidesLive Video
|
Soumajyoti Sarkar 路 Saab Mansour 路 Sailik Sengupta 路 Sheng Zha 路 Kaixiang Lin
馃敆
|
-
|
Depth-Wise Attention (DWAtt): A Layer Fusion Method for Data-Efficient Classification
(
Poster
)
>
|
Muhammad ElNokrashy 路 Badr AlKhamissi 路 Mona Diab
馃敆
|
-
|
SymbolicGPT: A Generative Transformer Model for Symbolic Regression
(
Poster
)
>
SlidesLive Video
|
Mojtaba Valipour 路 Bowen You 路 Maysum H Panju 路 Ali Ghodsi
馃敆
|
-
|
Using Informative Data Subsets for Efficient Training of Large Language Models: An Initial Study
(
Poster
)
>
SlidesLive Video
|
H S V N S Kowndinya Renduchintala 路 Krishnateja Killamsetty 路 Sumit Bhatia 路 Milan Aggarwal 路 Ganesh Ramakrishnan 路 Rishabh Iyer
馃敆
|
-
|
Using Selective Masking as a Bridge between Pre-training and Fine-tuning
(
Poster
)
>
SlidesLive Video
|
Tanish Lad 路 Himanshu Maheshwari 路 Shreyas Kottukkal 路 Radhika Mamidi
馃敆
|
-
|
Improved Knowledge Distillation by Utilizing Backward Pass Knowledge in Neural Networks
(
Poster
)
>
SlidesLive Video
|
Aref Jafari 路 Mehdi Rezaghoizadeh 路 Ali Ghodsi
馃敆
|
-
|
Topic Segmentation in the Wild: Towards Segmentation of Semi-structured & Unstructured Chats
(
Poster
)
>
SlidesLive Video
|
Reshmi Ghosh 路 Sharanya Kamath 路 Soundararajan Srinivasan 路 Dhuri Shrivastava 路 Samyadeep Basu 路 Harjeet Kajal
馃敆
|
-
|
A Theory of Unsupervised Translation for Understanding Animal Communication
(
Poster
)
>
SlidesLive Video
|
Shafi Goldwasser 路 David Gruber 路 Adam Tauman Kalai 路 Orr Paradise
馃敆
|
-
|
Collective Knowledge Graph Completion with Mutual Knowledge Distillation
(
Poster
)
>
SlidesLive Video
|
Weihang Zhang 路 Ovidiu Serban 路 Jiahao Sun 路 Yike Guo
馃敆
|
-
|
Gradient Knowledge Distillation for Pre-trained Language Models
(
Poster
)
>
SlidesLive Video
|
Lean Wang 路 Lei Li 路 Xu Sun
馃敆
|
-
|
Efficient Few-Shot Learning Without Prompts
(
Poster
)
>
SlidesLive Video
|
Oren Pereg 路 Daniel Korat 路 Moshe Wasserblat 路 Lewis Tunstall 路 Unso Eun Seo Jo 路 Luke Bates 路 Nils Reimers
馃敆
|
-
|
Fast DistilBERT on CPUs
(
Spotlight
)
>
|
Haihao Shen 路 Ofir Zafrir 路 Bo Dong 路 Hengyu Meng 路 Xinyu Ye 路 Zhe Wang 路 Yi Ding 路 Hanwen Chang 路 Guy Boudoukh 路 Moshe Wasserblat
馃敆
|
-
|
PCFG-based Natural Language Interface Improves Generalization for Controlled Text Generation
(
Spotlight
)
>
SlidesLive Video
|
Jingyu Zhang 路 Jim Glass 路 Tianxing He
馃敆
|
-
|
Improving the Robustness of DistilHuBERT to Unseen Noisy Conditions via Data Augmentation, Curriculum Learning, and Multi-Task Enhancement
(
Poster
)
>
|
Heitor Guimar茫es 路 Arthur Pimentel 路 Anderson R. Avila 路 Mehdi Rezaghoizadeh 路 Tiago H Falk
馃敆
|
-
|
Attribute Controlled Dialogue Prompting
(
Spotlight
)
>
|
Runcheng Liu 路 Ahmad Rashid 路 Ivan Kobyzev 路 Mehdi Rezaghoizadeh 路 Pascal Poupart
馃敆
|
-
|
PromptDA: Label-guided Data Augmentation for Prompt-based Few Shot Learners
(
Spotlight
)
>
SlidesLive Video
|
Canyu Chen 路 Kai Shu
馃敆
|
-
|
TBD7
(
KeyNote Talk
)
>
|
Kenneth Heafield
馃敆
|