Over 50 Mila-Affiliated Scientific Papers Accepted at ICLR 2024

Logo Mila and ICLR

From May 7 to May 11, 2024, dozens of Mila researchers will attend the twelfth International Conference on Learning Representations (ICLR 2024) in Vienna, Austria. This year, they will share 54 scientific papers at the main conference, showcasing their groundbreaking artificial intelligence (AI) research to peers from all around the world.

Here is a list of papers accepted at ICLR 2024 that contain at least one Mila-affiliated author:

TitleAuthorsPDF
Mastering Memory Tasks with World ModelsMohammad Reza Samsami, Artem Zholus, Janarthanan Rajendran, Sarath Chandarhttps://openreview.net/pdf?id=1vDArHJ68h
Course Correcting Koopman RepresentationsMahan Fathi, Clement Gehring, Jonathan Pilault, David Kanaa, Pierre-luc Bacon, Ross Goroshinhttps://openreview.net/pdf?id=A18gWgc5mi
Large Language Models as Generalizable Policies for Embodied TasksAndrew Szot, Max Schwarzer, Harsh Agrawal, Bogdan Mazoure, Rin Metcalf, Katherine Metcalf, Walter Talbott, Natalie Mackraz, Devon Hjelm, Alexander Toshevhttps://openreview.net/pdf?id=u6imHU4Ebu
Object-centric architectures enable efficient causal representation learningAmin Mansouri, Jason Hartford, Yan Zhang, Yoshua Bengiohttps://openreview.net/pdf?id=r9FsiXZxZt
On Diffusion Modeling for Anomaly DetectionVictor Livernoche, Vineet Jain, Yashar Hezaveh, Siamak Ravanbakhshhttps://openreview.net/pdf?id=lR3rk7ysXz
Synaptic Weight Distributions Depend on the Geometry of PlasticityRoman Pogodin, Jonathan Cornford, Arna Ghosh, Gauthier Gidel, Guillaume Lajoie, Blake Aaron Richardshttps://openreview.net/pdf?id=x5txICnnjC
Delta-AI: Local objectives for amortized inference in sparse graphical modelsJean-Pierre R. Falet, Hae Beom Lee, Nikolay Malkin, Chen Sun, Dragos Secrieru, Dinghuai Zhang, Guillaume Lajoie, Yoshua Bengiohttps://openreview.net/pdf?id=LemSSn8htt
Expected flow networks in stochastic environments and two-player zero-sum gamesMarco Jiralerspong, Bilun Sun, Danilo Vucetic, Tianyu Zhang, Yoshua Bengio, Gauthier Gidel, Nikolay Malkinhttps://openreview.net/pdf?id=uH0FGECSEI
Searching for High-Value Molecules Using Reinforcement Learning and TransformersRaj Ghugare, Santiago Miret, Adriana Hugessen, Mariano Phielipp, Glen Bersethhttps://openreview.net/pdf?id=O8mZO2ri33
Sufficient conditions for offline reactivation in recurrent neural networksNanda H Krishna, Colin Bredenberg, Daniel Levenstein, Blake Aaron Richards, Guillaume Lajoiehttps://openreview.net/pdf?id=RVrINT6MT7
Efficient Dynamics Modeling in Interactive Environments with Koopman TheoryArnab Kumar Mondal, Siba Smarak Panigrahi, Sai Rajeswar, K. Siddiqi, Siamak Ravanbakhshhttps://openreview.net/pdf?id=CmUWQ9s2D_
TACTiS-2: Better, Faster, Simpler Attentional Copulas for Multivariate Time SeriesArjun Ashok, Étienne Marcotte, Valentina Zantedeschi, Nicolas Chapados, Alexandre Drouinhttps://openreview.net/pdf?id=xtOydkE1Ku
Motif: Intrinsic Motivation from Artificial Intelligence FeedbackMartin Klissarov, P. D'Oro, Shagun Sodhani, Roberta Raileanu, Pierre-luc Bacon, Pascal Vincent, Amy Zhang, Mikael Henaffhttps://openreview.net/pdf?id=8v8AVAo6E5
On the Stability of Iterative Retraining of Generative Models on their own DataQuentin Bertrand, Avishek Joey Bose, Alexandre Duplessis, Marco Jiralerspong, Gauthier Gidelhttps://openreview.net/pdf?id=JORAfH2xFd
Leveraging Unpaired Data for Vision-Language Generative Models via Cycle ConsistencyTianhong Li, Sangnie Bhardwaj, Yonglong Tian, Han Zhang, Jarred Barber, Dina Katabi, Guillaume Lajoie, Huiwen Chang, Dilip Krishnanhttps://openreview.net/pdf?id=kNjrhD67LP
GOAt: Explaining Graph Neural Networks via Graph Output AttributionShengyao Lu, Keith G Mills, Jiao He, Bang Liu, Di Niuhttps://openreview.net/pdf?id=2Q8TZWAHv4
Improving Intrinsic Exploration by Creating Stationary ObjectivesRoger Creus Castanyer, Joshua Romoff, Glen Bersethhttps://openreview.net/pdf?id=YbZxT0SON4
Provable and Practical: Efficient Exploration in Reinforcement Learning via Langevin Monte CarloHaque Ishfaq, Qingfeng Lan, Pan Xu, A. Rupam Mahmood, Doina Precup, Anima Anandkumar, Kamyar Azizzadeneshelihttps://openreview.net/pdf?id=6u1z0RH6u1
Cycle Consistency Driven Object DiscoveryAniket Didolkar, Anirudh Goyal, Yoshua Bengiohttps://openreview.net/pdf?id=f1xnBr4WD6
Pre-Training and Fine-Tuning Generative Flow NetworksL. Pan, Moksh Jain, Kanika Madan, Yoshua Bengiohttps://openreview.net/pdf?id=2KY3WwgcTi
Towards Foundation Models for Knowledge Graph ReasoningMikhail Galkin, Xinyu Yuan, Hesham Mostafa, Jian Tang, Zhaocheng Zhuhttps://openreview.net/forum?id=jVEoydFOl9
PhyloGFN: Phylogenetic inference with generative flow networksMing Yang Zhou, Zichao Yan, Elliot Layne, Nikolay Malkin, Dinghuai Zhang, Moksh Jain, Mathieu Blanchette, Yoshua Bengiohttps://openreview.net/pdf?id=hB7SlfEmze
Reward Model Ensembles Help Mitigate OveroptimizationThomas Coste, Usman Anwar, Robert Kirk, David Kruegerhttps://openreview.net/pdf?id=NiQYQEPUsA
Consciousness-Inspired Spatio-Temporal Abstractions for Better Generalization in Reinforcement LearningMingde Zhao, Safa Alver, Harm van Seijen, Romain Laroche, Doina Precup, Yoshua Bengiohttps://openreview.net/pdf?id=eo9dHwtTFt
Amortizing intractable inference in large language modelsEdward J. Hu, Moksh Jain, Eric Elmoznino, Younesse Kaddar, Guillaume Lajoie, Yoshua Bengio, Nikolay Malkinhttps://openreview.net/pdf?id=Ouj6p4ca60
Piecewise Linear Parametrization of Policies: Towards Interpretable Deep Reinforcement LearningMaxime Wabartha, Joelle Pineauhttps://openreview.net/pdf?id=Zbt9z0a95l
Closing the Gap between TD Learning and Supervised Learning -- A Generalisation Point of View.Raj Ghugare, Matthieu Geist, Glen Berseth, Benjamin Eysenbachhttps://openreview.net/forum?id=qg5JENs0N4
Diffusion Generative Flow Samplers: Improving learning signals through partial trajectory optimizationDinghuai Zhang, Ricky T. Q. Chen, Cheng-Hao Liu, Aaron Courville, Yoshua Bengiohttps://openreview.net/pdf?id=OIsahq1UYC
Towards Foundational Models for Molecular Learning on Large-Scale Multi-Task DatasetsDominique Beaini, Shenyang Huang, Joao Alex Cunha, Zhiyi Li, Gabriela Moisescu-Pareja, Oleksandr Dymov, S. Maddrell-Mander, Callum McLean, Frederick Wenkel, Luis Müller, Jama Hussein Mohamud, Alipanah Parviz, Michael Craig, Michal Koziarski, Jiarui Lu, Zhaocheng Zhu, Cristian Gabellini, Kerstin Klaser, Josef Dean, Cas Wognum, Maciej Sypetkowski, Guillaume Rabusseau, Reihaneh Rabbany, Jian Tang, Ioannis Koutis, Christopher Morris, Mirco Ravanelli, Guy Wolf, Prudencio Tossou, Hadrien Mary, Thérence Bois, A. Fitzgibbon, Blazej Banaszewski, Chad Martin, Dominic Mastershttps://openreview.net/pdf?id=Zc2aIcucwc
Ghost on the Shell: An Expressive Representation of General 3D ShapesZhen Liu, Yao Feng, Yuliang Xiu, Weiyang Liu, Liam Paull, Michael J. Black, Bernhard Scholkopfhttps://openreview.net/pdf?id=Ad87VjRqUw
How connectivity structure shapes rich and lazy learning in neural circuitsYuhan Helena Liu, Aristide Baratin, Jonathan Cornford, Stefan Mihalas, Eric Shea-Brown, Guillaume Lajoiehttps://openreview.net/pdf?id=slSmYGc8ee
Würstchen: An Efficient Architecture for Large-Scale Text-to-Image Diffusion ModelsPablo Pernias, Dominic Rampas, Mats L. Richter, Christopher Joseph Pal, Marc Aubrevillehttps://openreview.net/pdf?id=gU58d5QeGv
Less or More From Teacher: Exploiting Trilateral Geometry For Knowledge DistillationChengming Hu, Haolun Wu, Xuan Li, Chen Ma, Xi Chen, Boyu Wang, Jun Yan, Xue Liuhttps://openreview.net/pdf?id=OZitfSXpdT
The Curse of Diversity in Ensemble-Based ExplorationZhixuan Lin, P. D'Oro, Evgenii Nikishin, Aaron Courvillehttps://openreview.net/pdf?id=M3QXCOTTk4
Poly-View Contrastive LearningAmitis Shidani, Dan Busbridge, Devon Hjelm, Jason Ramapuram, Eeshan Gunesh Dhekane, Russell Webbhttps://openreview.net/pdf?id=iHcTLIor0m
Improving Natural Language Understanding with Computation-Efficient Retrieval AugmentationShangyu Wu, Ying Xiong, Yufei CUI, Xue Liu, Buzhou Tang, Tei-Wei Kuo, Chun Jason Xuehttps://openreview.net/pdf?id=JtKGkz9fAe
Local Search GFlowNetsMinsu Kim, Taeyoung Yun, Emmanuel Bengio, Dinghuai Zhang, Yoshua Bengio, Sungsoo Ahn, Jinkyoo Parkhttps://openreview.net/pdf?id=6cFcw1Rxww
Balancing Act: Sparse Models with Constrained Disparate ImpactMeraj Hashemizadeh, Juan Ramirez, Rohan Sukumaran, Golnoosh Farnadi, Simon Lacoste-Julien, Jose Gallego-Posadahttps://openreview.net/pdf?id=Xz13DtbOVW
The Cost of Scaling Down Large Language Models: Reducing Model Size Affects Memory before In-context LearningTian Jin, Nolan Clement, X. Dong, Vaishnavh Nagarajan, Michael Carbin, Jonathan Ragan-Kelley, Gintare Karolina Dziugaitehttps://openreview.net/pdf?id=ldJXXxPE0L
Tree Cross AttentionLeo Feng, Frederick Tung, Hossein Hajimirsadeghi, Yoshua Bengio, M. O. Ahmedhttps://openreview.net/pdf?id=Vw24wtSddM
Decoupling regularization from the action spaceSobhan Mohammadpour, Emma Frejinger, Pierre-luc Baconhttps://openreview.net/pdf?id=UaMgmoKEBj
Bridging State and History Representations: Understanding Self-Predictive RLTianwei Ni, Benjamin Eysenbach, Erfan SeyedSalehi, Michel Ma, Clement Gehring, Aditya Mahajan, Pierre-luc Baconhttps://openreview.net/pdf?id=ms0VgzSGF2
LOQA: Learning with Opponent Q-Learning AwarenessMilad Aghajohari, Juan Agustin Duque, Tim Cooijmans, Aaron Courvillehttps://openreview.net/pdf?id=FDQF6A1s6M
Reasoning with Latent Diffusion in Offline Reinforcement LearningSiddarth Venkatraman, Shivesh Khaitan, Ravi Tej Akella, John Dolan, Jeff Schneider, Glen Bersethhttps://openreview.net/pdf?id=tGQirjzddO
Empirical Analysis of Model Selection for Heterogeneous Causal Effect EstimationDivyat Mahajan, Ioannis Mitliagkas, Brady Neal, Vasilis Syrgkanishttps://openreview.net/pdf?id=yuy6cGt3KL
Str2Str: A Score-based Framework for Zero-shot Protein Conformation SamplingJiarui Lu, Bozitao Zhong, Zuobai Zhang, Jian Tanghttps://openreview.net/pdf?id=C4BikKsgmK
What happens when you fine-tuning your model? Mechanistic analysis of procedurally generated tasks.Samyak Jain, Robert Kirk, E. S. Lubana, Robert P. Dick, Hidenori Tanaka, Tim Rocktäschel, Edward Grefenstette, David Kruegerhttps://openreview.net/pdf?id=A0HKeKl4Nl
Evaluating Representation Learning on the Protein Structure UniverseArian Rokkum Jamasb, Alex Morehead, Zuobai Zhang, Chaitanya K. Joshi, Kieran Didi, Simon V Mathis, Charles Harris, Jian Tang, Jianlin Cheng, Pietro Lio, Tom Leon Blundellhttps://openreview.net/pdf?id=sTYuRVrdK3
Intelligent Switching for Reset-Free RLDarshan Patil, Janarthanan Rajendran, Glen Berseth, Sarath Chandarhttps://openreview.net/pdf?id=Nq45xeghcL
Learning Multi-Agent Communication with Contrastive LearningYat Long Lo, Biswa Sengupta, Jakob Foerster, Michael Noukhovitchhttps://openreview.net/pdf?id=vZZ4hhniJU
SE(3)-Stochastic Flow Matching for Protein Backbone GenerationAvishek Joey Bose*, Tara Akhound-Sadegh*, Guillaume Huguet, Kilian FATRAS, Jarrid Rector-Brooks, Cheng-Hao Liu, Andrei Cristian Nica, Maksym Korablyov, Michael M. Bronstein, Alexander Tonghttps://openreview.net/pdf?id=kJFIH23hXb
Understanding In-Context Learning in Transformers and LLMs by Learning to Learn Discrete FunctionsSatwik Bhattamishra, Arkil Patel, Phil Blunsom, Varun Kanadehttps://openreview.net/pdf?id=ekeyCgeRfC
Ensemble Distillation for Unsupervised Constituency ParsingBehzad Shayegh, Yanshuai Cao, Xiaodan Zhu, Jackie CK Cheung, Lili Mouhttps://openreview.net/pdf?id=RR8y0WKrFv
GraphPulse: Topological representations for temporal graph property predictionKiarash Shamsi, Farimah Poursafaei, Shenyang Huang, Bao Tran Gia Ngo, Baris Coskunuzer, Cuneyt Gurcan Akcorahttps://openreview.net/pdf?id=DZqic2sPTY