164 citations · 175 across the 9 of their papers we have counts for
10 papers
On the Clock: Towards Punctual and Productive Time-Budgeted AI Agents
Aaron Wang, Neelabh Madan, Vlad Sobal +5
We study whether small LLM agents can operate effectively under explicit wall-clock time budgets by both respecting the allocated runtime and using available time productively. We…
StaminaBench: Stress-Testing Coding Agents over 100 Interaction Turns
Vlad Sobal, Shuo Yang, Yuting Zhang +2
We introduce StaminaBench, a benchmark that measures the stamina of coding agents: how many consecutive interaction turns (change requests) they can handle before failing. Unlike t…
Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models
Vlad Sobal, Wancong Zhang, Kyunghyun Cho +3
A long-standing goal in AI is to develop agents capable of solving diverse tasks across a range of environments, including those never seen during training. Two dominant paradigms…
-Sample Contrastive Loss: Improving Contrastive Learning with Sample Similarity Graphs
Vlad Sobal, Mark Ibrahim, Randall Balestriero +5
Learning good representations involves capturing the diverse ways in which data samples relate. Contrastive loss - an objective matching related samples - underlies methods from se…
Hierarchical World Models as Visual Whole-Body Humanoid Controllers
Nicklas Hansen, Jyothir S, Vlad Sobal +3
Whole-body control for humanoids is challenging due to the high-dimensional nature of the problem, coupled with the inherent instability of a bipedal morphology. Learning from visu…
Gradient-based Planning with World Models
Jyothir S, Siddhartha Jalagam, Yann LeCun +1
The enduring challenge in the field of artificial intelligence has been the control of systems to achieve desired behaviours. While for systems governed by straightforward dynamics…
A Cookbook of Self-Supervised Learning
Randall Balestriero, Mark Ibrahim, Vlad Sobal +16
Self-supervised learning, dubbed the dark matter of intelligence, is a promising path to advance machine learning. Yet, much like cooking, training SSL methods is a delicate art wi…
Joint Embedding Predictive Architectures Focus on Slow Features
Vlad Sobal, Jyothir S, Siddhartha Jalagam +3
Many common methods for learning a world model for pixel-based environments use generative architectures trained with pixel-level reconstruction objectives. Recently proposed Joint…
Light-weight probing of unsupervised representations for Reinforcement Learning
Wancong Zhang, Anthony GX-Chen, Vlad Sobal +2
Unsupervised visual representation learning offers the opportunity to leverage large corpora of unlabeled trajectories to form useful visual representations, which can benefit the…
Separating the World and Ego Models for Self-Driving
Vlad Sobal, Alfredo Canziani, Nicolas Carion +2
Training self-driving systems to be robust to the long-tail of driving scenarios is a critical problem. Model-based approaches leverage simulation to emulate a wide range of scenar…