collaborators

15 papers

cs.LG2026

ExpRL: Exploratory RL for LLM Mid-Training

Violet Xiang, Amrith Setlur, Chase Blagden +2

Sparse reward reinforcement learning (RL) has become a standard tool for improving LLM reasoning, but its success depends critically on the coverage present in the base model. In p…

cs.AI2026

QED-Nano: Teaching a Tiny Model to Prove Hard Theorems

LM-Provers, Yuxiao Qu, Amrith Setlur +6

Proprietary AI systems have recently demonstrated impressive capabilities on complex proof-based problems, with gold-level performance reported at the 2025 International Mathematic…

cs.LG2026

Reasoning Cache: Continual Improvement Over Long Horizons via Short-Horizon RL

Ian Wu, Yuxiao Qu, Amrith Setlur +1

Large Language Models (LLMs) that can continually improve beyond their training budgets are able to solve increasingly difficult problems by adapting at test time, a property we re…

cs.LG2026

IsoCompute Playbook: Optimally Scaling Sampling Compute for LLM RL

Zhoujun Cheng, Yutao Xie, Yuxiao Qu +12

While scaling laws guide compute allocation for LLM pre-training, analogous prescriptions for reinforcement learning (RL) post-training of large language models (LLMs) remain poorl…

cs.LG2026

Reuse your FLOPs: Scaling RL on Hard Problems by Conditioning on Very Off-Policy Prefixes

Amrith Setlur, Zijian Wang, Andrew Cohen +2

Typical reinforcement learning (RL) methods for LLM reasoning waste compute on hard problems, where correct on-policy traces are rare, policy gradients vanish, and learning stalls.…

cs.LG2026

POPE: Learning to Reason on Hard Problems via Privileged On-Policy Exploration

Yuxiao Qu, Amrith Setlur, Virginia Smith +2

Reinforcement learning (RL) has improved the reasoning abilities of large language models (LLMs), yet state-of-the-art methods still fail to learn on many training problems. On har…