9 papers · 1 filter
LeanPRM: Lean-Grounded Process Supervision for Natural Language Proof Verification
Rodney Lafuente-Mercado, Salman Rahman, Jingyan Shen +3
AI systems now write long and difficult mathematical proofs, and checking their step-by-step validity is becoming a bottleneck. A learned verifier could check proofs cheaply, but t…
Locating Hidden Failures Makes Long-Horizon Agents More Reliable
Salman Rahman, Yubin Kim, Mihir Parmar +15
As AI agents take on long, autonomous tasks, we increasingly oversee rather than perform the work, yet we still judge them almost entirely by whether they finally succeed. An outco…
Understanding Reasoning from Pretraining to Post-Training
Jingyan Shen, Ang Li, Salman Rahman +4
Reinforcement learning (RL) has become central to improving large language models (LLMs) on complex reasoning tasks, yet RL post-training is largely studied in isolation from the p…
How's it going? Reinforcement learning in language models recruits a functional welfare axis
Andy Q Han, David J. Chalmers, Pavel Izmailov
How does reinforcement learning shape a language model's internal representations? We present evidence that RL recruits a pre-existing representation of functional welfare: an esti…
Forgetting in Language Models: Capacity, Optimization, and Self-Generated Replay
Martin Marek, Dongkyu Cho, Shikai Qiu +3
Models trained on a new task typically degrade on prior tasks, a phenomenon known as forgetting. Traditionally, mitigating forgetting has required replaying stored exemplars from p…
When Can LLMs Learn to Reason with Weak Supervision?
Salman Rahman, Jingyan Shen, Anna Mordvina +3
Large language models have achieved significant reasoning improvements through reinforcement learning with verifiable rewards (RLVR). Yet as model capabilities grow, constructing h…
Reliable and Responsible Foundation Models: A Comprehensive Survey
Xinyu Yang, Junlin Han, Rishi Bommasani +49
Foundation models, including Large Language Models (LLMs), Multimodal Large Language Models (MLLMs), Image Generative Models (i.e, Text-to-Image Models and Image-Editing Models), a…
Closing the Train-Test Gap in World Models for Gradient-Based Planning
Arjun Parthasarathy, Nimit Kalra, Rohun Agrawal +4
World models paired with model predictive control (MPC) can be trained offline on large-scale datasets of expert trajectories and enable generalization to a wide range of planning…
Can a Confident Prior Replace a Cold Posterior?
Martin Marek, Brooks Paige, Pavel Izmailov
Benchmark datasets used for image classification tend to have very low levels of label noise. When Bayesian neural networks are trained on these datasets, they often underfit, misr…