Workshop
Robustness in Sequence Modeling
Nathan Ng 路 Haoran Zhang 路 Vinith Suriyakumar 路 Chantal Shaib 路 Kyunghyun Cho 路 Yixuan Li 路 Alice Oh 路 Marzyeh Ghassemi
Room 290
Fri 2 Dec, 7 a.m. PST
As machine learning models find increasing use in the real world, ensuring their safe and reliable deployment depends on ensuring their robustness to distribution shift. This is especially true for sequential data, which occurs naturally in various data domains such as natural language processing, healthcare, computational biology, and finance. However, building models for sequence data which are robust to distribution shifts presents a unique challenge. Sequential data are often discrete rather than continuous, exhibit difficult to characterize distributions, and can display a much greater range of types of distributional shifts. Although many methods for improving model robustness exist for imaging or tabular data, extending these methods to sequential data is a challenging research direction that often requires fundamentally different techniques.
This workshop aims to facilitate progress towards improving the distributional robustness of models trained on sequential data by bringing together researchers to tackle a wide variety of research questions including, but not limited to:
(1) How well do existing robustness methods work on sequential data, and why do they succeed or fail?
(2) How can we leverage the sequential nature of the data to develop novel and distributionally robust methods?
(3) How do we construct and utilize formalisms for distribution shifts in sequential data?
We hope that this workshop provides a first step towards improving the robustness, and ultimately safety and reliability, of models in sequential data domains.
Schedule
Fri 7:00 a.m. - 7:15 a.m.
|
Opening Remarks
(
Opening Remarks
)
>
SlidesLive Video |
馃敆 |
Fri 7:15 a.m. - 8:00 a.m.
|
Invited Talk: Behnam Neyshabur
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 8:00 a.m. - 8:10 a.m.
|
An Invariant Learning Characterization of Controlled Text Generation
(
Spotlight
)
>
link
SlidesLive Video |
Claudia Shi 路 Carolina Zheng 路 Keyon Vafa 路 Amir Feder 路 David Blei 馃敆 |
Fri 8:10 a.m. - 8:20 a.m.
|
Exploiting Variable Correlation with Masked Modeling for Anomaly Detection in Time Series
(
Spotlight
)
>
link
SlidesLive Video |
Panagiotis Lymperopoulos 路 Yukun Li 路 Liping Liu 馃敆 |
Fri 8:20 a.m. - 8:30 a.m.
|
Online Policy Optimization for Robust MDP
(
Spotlight
)
>
link
SlidesLive Video |
Jing Dong 路 Jingwei Li 路 Baoxiang Wang 路 Jingzhao Zhang 馃敆 |
Fri 8:30 a.m. - 9:30 a.m.
|
Poster Session #1
(
Poster Session
)
>
|
馃敆 |
Fri 9:30 a.m. - 10:00 a.m.
|
Invited Talk: Dan Hendrycks
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 10:00 a.m. - 10:30 a.m.
|
Invited Talk: Suchi Saria
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 10:30 a.m. - 11:30 a.m.
|
Lunch Break
|
馃敆 |
Fri 11:30 a.m. - 12:15 p.m.
|
Invited Talk: Yejin Choi
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 12:15 p.m. - 12:25 p.m.
|
Comparison of Uncertainty Quantification in Time Series Regression
(
Spotlight
)
>
link
SlidesLive Video |
Levente Foldesi 路 Matias Valdenegro-Toro 馃敆 |
Fri 12:25 p.m. - 12:35 p.m.
|
Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors ( Spotlight ) > link | Thomas Hartvigsen 路 Swami Sankaranarayanan 路 Hamid Palangi 路 Yoon Kim 路 Marzyeh Ghassemi 馃敆 |
Fri 12:35 p.m. - 12:45 p.m.
|
Out-of-Distribution Detection and Selective Generation for Conditional Language Models
(
Spotlight
)
>
link
SlidesLive Video |
Jie Ren 路 Jiaming Luo 路 Yao Zhao 路 Kundan Krishna 路 Mohammad Saleh 路 Balaji Lakshminarayanan 路 Peter Liu 馃敆 |
Fri 12:45 p.m. - 1:45 p.m.
|
Poster Session #2
(
Poster Session
)
>
|
馃敆 |
Fri 1:45 p.m. - 2:15 p.m.
|
Invited Talk: Byron Wallace
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 2:15 p.m. - 2:45 p.m.
|
Invited Talk: He He
(
Invited Talk
)
>
SlidesLive Video |
馃敆 |
Fri 2:45 p.m. - 3:00 p.m.
|
Closing Remarks
(
Closing Remarks
)
>
SlidesLive Video |
馃敆 |
-
|
Perturbation Augmentation for Fairer NLP ( Poster ) > link | Rebecca Qian 路 Candace Ross 路 Jude Fernandes 路 Eric Michael Smith 路 Douwe Kiela 路 Adina Williams 馃敆 |
-
|
Aging with GRACE: Lifelong Model Editing with Discrete Key-Value Adaptors ( Poster ) > link | Thomas Hartvigsen 路 Swami Sankaranarayanan 路 Hamid Palangi 路 Yoon Kim 路 Marzyeh Ghassemi 馃敆 |
-
|
Feature Restricted Group Dropout for Robust Electronic Health Record Predictions ( Poster ) > link | Bret Nestor 路 Anna Goldenberg 路 Marzyeh Ghassemi 馃敆 |
-
|
CLIFT : Analysing Natural Distribution Shift on Question Answering Models in Clinical Domain ( Poster ) > link | Ankit Pal 馃敆 |
-
|
Probabilistic thermal stability prediction through sparsity promoting transformer representation ( Poster ) > link | Yevgen Zainchkovskyy 路 Jesper Ferkinghoff-Borg 路 Anja Bennett 路 Thomas Egebjerg 路 Nikolai Lorenzen 路 Per Greisen 路 S酶ren Hauberg 路 Carsten Stahlhut 馃敆 |
-
|
On the Abilities of Sequence Extrapolation with Implicit Models ( Poster ) > link | Juliette Decugis 路 Alicia Tsai 路 Ashwin Ganesh 路 Max Emerling 路 Laurent El Ghaoui 馃敆 |
-
|
An Invariant Learning Characterization of Controlled Text Generation ( Poster ) > link | Claudia Shi 路 Carolina Zheng 路 Keyon Vafa 路 Amir Feder 路 David Blei 馃敆 |
-
|
Robustness of Neural Networks used in Electrical Motor Time-Series ( Poster ) > link | Sagar Verma 路 Kavya Gupta 馃敆 |
-
|
Quantifying Uncertainty in Foundation Models via Ensembles ( Poster ) > link | Meiqi Sun 路 Wilson Yan 路 Pieter Abbeel 路 Igor Mordatch 馃敆 |
-
|
An Adaptive Temporal Attention Mechanism to Address Distribution Shifts ( Poster ) > link | sepideh koohfar 路 Laura Dietz 馃敆 |
-
|
Defend Against Textual Backdoor Attacks By Token Substitution ( Poster ) > link | Xinglin Li 路 Yao Li 路 Minhao Cheng 馃敆 |
-
|
Behavioral Classification of Sequential Neural Activity Using Time Varying Recurrent Neural Networks ( Poster ) > link | Yongxu Zhang 路 Shreya Saxena 馃敆 |
-
|
Strategy-Aware Contextual Bandits ( Poster ) > link | Keegan Harris 路 Chara Podimata 路 Steven Wu 馃敆 |
-
|
Revealing the Bias in Large Language Models via Reward Structured Questions ( Poster ) > link | Ezgi Korkmaz 馃敆 |
-
|
Online Policy Optimization for Robust MDP ( Poster ) > link | Jing Dong 路 Jingwei Li 路 Baoxiang Wang 路 Jingzhao Zhang 馃敆 |
-
|
Exploiting Variable Correlation with Masked Modeling for Anomaly Detection in Time Series ( Poster ) > link | Panagiotis Lymperopoulos 路 Yukun Li 路 Liping Liu 馃敆 |
-
|
Comparison of Uncertainty Quantification in Time Series Regression ( Poster ) > link | Levente Foldesi 路 Matias Valdenegro-Toro 馃敆 |
-
|
Out-of-Distribution Detection and Selective Generation for Conditional Language Models ( Poster ) > link | Jie Ren 路 Jiaming Luo 路 Yao Zhao 路 Kundan Krishna 路 Mohammad Saleh 路 Balaji Lakshminarayanan 路 Peter Liu 馃敆 |
-
|
Are Deep Sequence Classifiers Good at Non-Trivial Generalization? ( Poster ) > link | Francesco Cazzaro 路 Ariadna Quattoni 路 Xavier Carreras 馃敆 |
-
|
Conditional COT-GAN for Video Prediction with Kernel Smoothing ( Poster ) > link | Tianlin Xu 路 Beatrice Acciaio 馃敆 |