Publications
Articles de conférence et de revue
2024
-
Balancing Context Length and Mixing Times for Reinforcement Learning at Scale
Matthew Riemer, Khimya Khetarpal, Janarthanan Rajendran et Sarath Chandar
Neural Information Processing Systems (NeurIPS), 2024.
#RL
-
Exploring Quantization for Efficient Pre-Training of Transformer Language Models
Kamran Chitsaz, Quentin Fournier, Gonçalo Mordido et Sarath Chandar
Findings of the Association for Computational Linguistics (EMNLP), 2024.
#NLP, #DL
[arXiv] -
Context-Aware Assistant Selection for Improved Inference Acceleration with Large Language Models
Jerry Huang, Prasanna Parthasarathi, Mehdi Rezagholizadeh et Sarath Chandar
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024.
#NLP
[arXiv] -
Do Large Language Models Know How Much They Know?
Gabriele Prato, Jerry Huang, Prasanna Parthasarathi, Shagun Sodhani et Sarath Chandar
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2024.
#NLP
-
Toward Debugging Deep Reinforcement Learning Programs with RLExplorer
Rached Bouchoucha, Ahmed Haj Yahmed, Darshan Patil, Janarthanan Rajendran, Amin Nikanjam, Sarath Chandar et Foutse Khomh
International Conference on Software Maintenance and Evolution (ICSME), 2024.
#RL
-
WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Tasks
Leo Boisvert*, Megh Thakkar*, Maxime Gasse, Massimo Caccia, Thibault Le Sellier De Chezelles, Quentin Cappart, Nicolas Chapados, Alexandre Lacoste et Alexandre Drouin
Neural Information Processing Systems (NeurIPS), Datasets and Benchmarks, 2024.
#NLP
[arXiv], [code] -
Should We Attend More or Less? Modulating Attention for Fairness
Abdelrahman Zayed, Gonçalo Mordido, Samira Shabanian et Sarath Chandar
Conference on Language Modeling (COLM), 2024.
#NLP
[arXiv] -
Are self-explanations from Large Language Models faithful?
Andreas Madsen, Sarath Chandar et Siva Reddy
Findings of the Association for Computational Linguistics (ACL), 2024.
#NLP
[arXiv], [code], [YouTube] -
A deep-dive into the tradeoffs of preference alignment with PEFT
Megh Thakkar, Quentin Fournier, Matthew Riemer, Pin-Yu Chen, Amal Zouaq, Payel Das et Sarath Chandar
Annual Meeting of the Association for Computational Linguistics (ACL), 2024.
#NLP
[arXiv] -
Why Don’t Prompt-Based Fairness Metrics Correlate?
Abdelrahman Zayed, Gonçalo Mordido, Ioana Baldini et Sarath Chandar
Annual Meeting of the Association for Computational Linguistics (ACL), 2024.
#NLP
[arXiv], [YouTube] -
Sub-goal Distillation: A Method to Improve Small Language Agents
Maryam Hashemzadeh, Elias Stengel-Eskin, Sarath Chandar et Marc-Alexandre Cote
Conference on Lifelong Learning Agents (CoLLAs), 2024. [Oral presentation.]
#RL, #NLP
[arXiv] -
Partial Models for Building Adaptive Model-Based Reinforcement Learning Agents
Safa Alver, Ali Rahimi-Kalahroudi et Doina Precup
Conference on Lifelong Learning Agents (CoLLAs), 2024.
#RL
[arXiv] -
Lookbehind-SAM: k steps back, 1 step forward
Gonçalo Mordido, Pranshu Malviya, Aristide Baratin et Sarath Chandar
International Conference on Machine Learning (ICML), 2024.
#DL
[arXiv], [code], [YouTube] -
Faithfulness Measurable Masked Language Models
Andreas Madsen, Siva Reddy et Sarath Chandar
International Conference on Machine Learning (ICML), 2024. [Spotlight award - top 3.5%]
#NLP
[arXiv], [code], [YouTube], [blogpost] -
Promoting Exploration in Memory-Augmented Adam using Critical Momenta
Pranshu Malviya, Gonçalo Mordido, Aristide Baratin, Reza Babanezhad Harikandeh, Jerry Huang, Simon Lacoste-Julien, Razvan Pascanu et Sarath Chandar
Transactions on Machine Learning Research (TMLR), 2024.
#DL
[arXiv] -
A Responsible Framework for Applying Artificial Intelligence on Medical Images and Signals at the Point-of-care: the PACS-AI Platform
Pascal Theriault-Lauzier, Denis Cobin, Olivier Tastet, Elodie Labrecque Langlais, Bahareh Taji, Guson Kang, Aun-Yeong Chong, Derek So, An Tang, Judy Wawira Gichoya, Sarath Chandar, Pierre-Luc Déziel, Julie G Hussin, Samuel Kadoury et Robert Avram
Canadian Journal of Cardiology, 2024.
#DL, #Other
-
MVP: Minimal Viable Phrase for Long Text Understanding
Louis Clouâtre, Amal Zouaq et Sarath Chandar
Joint International Conference on Computational Linguistics, Language, Resources and Evaluation (LREC-COLING), 2024.
#NLP
-
Mastering Memory Tasks with World Models
Mohammad Reza Samsami*, Artem Zholus*, Janarthanan Rajendran et Sarath Chandar
International Conference on Learning Representations (ICLR), 2024. [Oral presentation.]
#RL, #DL
[openreview] -
Intelligent Switching for Reset-Free RL
Darshan Patil, Janarthanan Rajendran, Glen Berseth et Sarath Chandar
International Conference on Learning Representations (ICLR), 2024.
#RL
[openreview] -
On the Costs and Benefits of Adopting Lifelong Learning for Software Analytics - Empirical Study on Brown Build and Risk Prediction
Doriane Olewicki, Sarra Habchi, Mathieu Nayrolles, Mojtaba Faramarzi, Sarath Chandar et Bram Adams
International Conference on Software Engineering (ICSE) - Software Engineering in Practice Track, 2024. [ICSE24 SEIP Distinguished Paper Award.]
#DL
[arXiv] -
Fast and Accurate Output Error Estimation for Memristor-Based Deep Neural Networks
Jonathan Kern, Sébastien Henwood, Gonçalo Mordido, Elsa Dupraz, Abdeldjalil Aïssa-El-Bey, Yvon Savaria et François Leduc-Primeau
IEEE Transactions on Signal Processing, 2024.
#DL
[paper] -
Fairness-Aware Structured Pruning in Transformers
Abdelrahman Zayed, Gonçalo Mordido, Samira Shabanian, Ioana Baldini et Sarath Chandar
AAAI Conference on Artificial Intelligence (AAAI), 2024.
#NLP
[arXiv], [YouTube] -
Learning Conditional Policies for Crystal Design Using Offline Reinforcement Learning
Prashant Govindarajan, Santiago Miret, Jarrid Rector-Brooks, Mariano Phielipp, Janarthanan Rajendran et Sarath Chandar
Digital Discovery Journal, 2024.
#RL
[openreview]
2023
-
Self-Influence Guided Data Reweighting for Language Model Pre-training
Megh Thakkar, Tolga Bolukbasi, Sriram Ganapathy, Shikhar Vashishth, Sarath Chandar et Partha Talukdar
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023.
#NLP
[arXiv] -
EpiK-Eval: Evaluation for Language Models as Epistemic Models
Gabriele Prato, Jerry Huang, Prasanna Parthasarathi, Shagun Sodhani et Sarath Chandar
Conference on Empirical Methods in Natural Language Processing (EMNLP), 2023.
#NLP
[arXiv], [code] -
Measuring the Knowledge Acquisition-Utilization Gap in Pretrained Language Models
Amirhossein Kazemnejad, Mehdi Rezagholizadeh, Prasanna Parthasarathi et Sarath Chandar
Findings of the Association for Computational Linguistics (EMNLP), 2023.
#NLP
[arXiv] -
Training DNNs Resilient to Adversarial and Random Bit-Flips by Learning Quantization Ranges
Kamran Chitsaz, Gonçalo Mordido, Jean Pierre David et François Leduc-Primeau
Transactions on Machine Learning Research (TMLR), 2023.
#DL
[openreview], [code] -
Replay Buffer with Local Forgetting for Adapting to Local Environment Changes in Deep Model-Based Reinforcement Learning
Ali Rahimi-Kalahroudi, Janarthanan Rajendran, Ida Momennejad, Harm van Seijen et Sarath Chandar
Conference on Lifelong Learning Agents (CoLLAs), 2023.
[Deep Reinforcement Learning Workshop, NeurIPS, 2022]
#RL
[arXiv] -
Towards Few-shot Coordination: Revisiting Ad-hoc Teamplay Challenge In the Game of Hanabi
Hadi Nekoei, Xutong Zhao, Janarthanan Rajendran, Miao Liu et Sarath Chandar
Conference on Lifelong Learning Agents (CoLLAs), 2023.
#RL
[paper] -
Dealing With Non-stationarity in Decentralized Cooperative Multi-Agent Deep Reinforcement Learning via Multi-Timescale Learning
Hadi Nekoei, Akilesh Badrinaaraayanan, Amit Sinha, Mohammad Amini, Janarthanan Rajendran, Aditya Mahajan et Sarath Chandar
Conference on Lifelong Learning Agents (CoLLAs), 2023.
#RL
[arXiv] -
Conditionally Optimistic Exploration for Cooperative Deep Multi-Agent Reinforcement Learning
Xutong Zhao, Yangchen Pan, Chenjun Xiao, Sarath Chandar et Janarthanan Rajendran
Conference on Uncertainty in Artificial Intelligence (UAI), 2023.
#RL
[arXiv] -
An Empirical Investigation of the Role of Pre-training in Lifelong Learning
Sanket Vaibhav Mehta, Darshan Patil, Sarath Chandar et Emma Strubell
Journal of Machine Learning Research, 2023.
#DL
[arXiv] -
Multi-Agent Reinforcement Learning for Fast-Timescale Demand Response of Residential Loads
Vincent Mai, Philippe Maisonneuve, Tianyu Zhang, Hadi Nekoei, Liam Paull et Antoine Lesage-Landry
International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 2023.
#RL
[arXiv] -
Deep Learning on a Healthy Data Diet: Finding Important Examples for Fairness
Abdelrahman Zayed, Prasanna Parthasarathi, Gonçalo Mordido, Hamid Palangi, Samira Shabanian et Sarath Chandar
AAAI Conference on Artificial Intelligence (AAAI), 2023.
#NLP
[arXiv], [YouTube] -
DEUP: Direct Epistemic Uncertainty Prediction
Moksh Jain, Salem Lahlou, Hadi Nekoei, Victor Butoi, Paul Bertin, Jarrid Rector-Brooks, Maksym Korablyov et Yoshua Bengio
Transactions on Machine Learning Research (TMLR), 2023.
#DL
[arXiv], [code] -
Label fusion and training methods for reliable representation of inter-rater uncertainty
Andreanne Lemay, Charley Gros, Naga Karthik Enamundram et Julien Cohen-Adad
The Journal of Machine Learning for Biomedical Imaging (MELBA), 2023.
#DL
[paper]
2022
-
Evaluating the Faithfulness of Importance Measures in NLP by Recursively Masking Allegedly Important Tokens and Retraining
Andreas Madsen, Nicholas Meade, Vaibhav Adlakha et Siva Reddy
Findings of the Association for Computational Linguistics (EMNLP), 2022.
[BlackboxNLP Workshop, 2022]
#NLP
[arXiv], [code] -
Detecting Languages Unintelligible to Multilingual Models through Local Structure Probes
Louis Clouâtre, Prasanna Parthasarathi, Amal Zouaq et Sarath Chandar
Findings of the Association for Computational Linguistics (EMNLP), 2022.
#NLP
-
Local Structure Matters Most in Most Languages
Louis Clouâtre, Prasanna Parthasarathi, Amal Zouaq et Sarath Chandar
Conference of the Asia-Pacific Chapter of the Association for Computational Linguistics and the International Joint Conference on Natural Language Processing (AACL-IJCNLP), 2022.
#NLP
-
TAG: Task-based Accumulated Gradients for Lifelong Learning
Pranshu Malviya, Balaraman Ravindran et Sarath Chandar
Conference on Lifelong Learning Agents (CoLLAs), 2022.
[Workshop on Theory and Foundation of Continual Learning, ICML, 2021]
#DL
[arXiv], [code] -
Improving Meta-Learning Generalization with Activation-Based Early-Stopping
Simon Guiroy, Christopher Pal, Gonçalo Mordido et Sarath Chandar
Conference on Lifelong Learning Agents (CoLLAs), 2022.
#DL
[arXiv], [code], [YouTube] -
Combining Reinforcement Learning and Constraint Programming for Sequence-Generation Tasks with Hard Constraints
Daphné Lafleur, Sarath Chandar et Gilles Pesant
Principles and Practice of Constraint Programming (CP), 2022.
#RL
-
Biological Sequence Design with GFlowNets
Moksh Jain, Emmanuel Bengio, Alex-Hernandez Garcia, Jarrid Rector-Brooks, Bonaventure F. P. Dossou, Chanakya Ekbote, Jie Fu, Tianyu Zhang, Micheal Kilgour, Dinghuai Zhang, Lena Simine, Payel Das et Yoshua Bengio
International Conference on Machine Learning (ICML), 2022.
#DL
[arXiv], [code] -
Towards Evaluating Adaptivity of Model-Based Reinforcement Learning Methods
Yi Wan*, Ali Rahimi-Kalahroudi*, Janarthanan Rajendran, Ida Momennejad, Sarath Chandar et Harm van Seijen
International Conference on Machine Learning (ICML), 2022.
#RL
[arXiv], [code] -
Post-hoc Interpretability for Neural NLP: A Survey
Andreas Madsen, Siva Reddy et Sarath Chandar
ACM Computing Surveys, 2022.
#NLP
[arXiv] -
Local Structure Matters Most: Perturbation Study in NLU
Louis Clouâtre, Prasanna Parthasarathi, Amal Zouaq et Sarath Chandar
Findings of the Association for Computational Linguistics (ACL), 2022.
#NLP
[arXiv] -
Memory Augmented Optimizers for Deep Learning
Paul-Aymeric McRae, Prasanna Parthasarathi, Mido Assran et Sarath Chandar
International Conference on Learning Representations (ICLR), 2022.
#DL
[openreview], [code] -
PatchUp: A Feature-Space Block-Level Regularization Technique for Convolutional Neural Networks
Mojtaba Faramarzi, Mohammad Amini, Akilesh Badrinaaraayanan, Vikas Verma et Sarath Chandar
AAAI Conference on Artificial Intelligence (AAAI), 2022.
#DL
[arXiv], [code]
2021
-
MLMLM: Link Prediction with Mean Likelihood Masked Language Model
Louis Clouâtre, Philippe Trempe, Amal Zouaq et Sarath Chandar
Findings of the Association for Computational Linguistics (ACL-IJCNLP), 2021.
#NLP
[arXiv] -
Benchmarking Bias Mitigation Algorithms in Representation Learning through Fairness Metrics
Charan Reddy, Deepak Sharma, Soroush Mehri, Adriana Romero, Samira Shabanian et Sina Honari
Proceedings of the Neural Information Processing Systems (NeurIPS) Track on Datasets and Benchmarks, 2021.
#NLP
[openreview], [code] -
A Brief Study on the Effects of Training Generative Dialogue Models with a Semantic loss
Prasanna Parthasarathi, Mohamed Abdelsalam, Joelle Pineau et Sarath Chandar
Proceedings of the 22nd Annual SIGdial Meeting on Discourse and Dialogue, 2021.
#NLP
-
Do Encoder Representations of Generative Dialogue Models Encode Sufficient Information about the Task ?
Prasanna Parthasarathi, Sarath Chandar et Joelle Pineau
Proceedings of the 22nd Annual SIGdial Meeting on Discourse and Dialogue, 2021.
#NLP
-
Continuous Coordination As a Realistic Scenario for Lifelong Learning
Hadi Nekoei, Akilesh Badrinaaraayanan, Aaron Courville et Sarath Chandar
International Conference on Machine Learning (ICML), 2021.
#RL
[arXiv], [code] -
A Survey of Data Augmentation Approaches for NLP
Steven Y. Feng, Varun Gangal, Jason Wei, Sarath Chandar, Soroush Vosoughi, Teruko Mitamura et Eduard Hovy
Findings of the Association for Computational Linguistics (ACL-IJCNLP), 2021.
#NLP
[arXiv] -
Towered Actor Critic for Handling Multiple Action Types in Reinforcement Learning For Drug Discovery
Sai Krishna Gottipati, Yashaswi Pathak, Boris Sattarov, Sahir, Rohan Nuttall, Mohammad Amini, Matthew E. Taylor et Sarath Chandar
AAAI Conference on Artificial Intelligence (AAAI), 2021.
#RL
-
IIRC: Incremental Implicitly-Refined Classification
Mohamed Abdelsalam, Mojtaba Faramarzi, Shagun Sodhani et Sarath Chandar
Conference on Computer Vision and Pattern Recognition (CVPR), 2021.
#DL
[arXiv], [code], [website], [PyPI], [docs]
2020
-
The LoCA Regret: A Consistent Metric to Evaluate Model-Based Behavior in Reinforcement Learning
Harm van Seijen, Hadi Nekoei, Evan Racah et Sarath Chandar
Neural Information Processing Systems (NeurIPS), 2020.
#RL
[arXiv], [code] -
Learning To Navigate The Synthetically Accessible Chemical Space Using Reinforcement Learning
Sai Krishna Gottipati*, Boris Sattarov*, Sufeng Niu, Yashaswi Pathak, Haoran Wei, Shengchao Liu, Karam MJ Thomas, Simon Blackburn, Connor W Coley, Jian Tang, Sarath Chandar et Yoshua Bengio
International Conference on Machine Learning (ICML), 2020.
#RL
[arXiv]