Picture for Michael Rabbat

Michael Rabbat

McGill University, Montreal, Canada

V-JEPA 2: Self-Supervised Video Models Enable Understanding, Prediction and Planning

Add code
Jun 11, 2025
Viaarxiv icon

IntPhys 2: Benchmarking Intuitive Physics Understanding In Complex Synthetic Environments

Add code
Jun 11, 2025
Viaarxiv icon

Locate 3D: Real-World Object Localization via Self-Supervised Learning in 3D

Add code
Apr 19, 2025
Viaarxiv icon

Scaling Language-Free Visual Representation Learning

Add code
Apr 01, 2025
Viaarxiv icon

Accelerating Neural Network Training: An Analysis of the AlgoPerf Competition

Add code
Feb 20, 2025
Viaarxiv icon

Intuitive physics understanding emerges from self-supervised pretraining on natural videos

Add code
Feb 17, 2025
Viaarxiv icon

Towards General-Purpose Model-Free Reinforcement Learning

Add code
Jan 27, 2025
Figure 1 for Towards General-Purpose Model-Free Reinforcement Learning
Figure 2 for Towards General-Purpose Model-Free Reinforcement Learning
Figure 3 for Towards General-Purpose Model-Free Reinforcement Learning
Figure 4 for Towards General-Purpose Model-Free Reinforcement Learning
Viaarxiv icon

MetaMorph: Multimodal Understanding and Generation via Instruction Tuning

Add code
Dec 18, 2024
Viaarxiv icon

Dualformer: Controllable Fast and Slow Thinking by Learning with Randomized Reasoning Traces

Add code
Oct 13, 2024
Viaarxiv icon

Embracing Diversity: Interpretable Zero-shot classification beyond one vector per class

Add code
Apr 25, 2024
Viaarxiv icon