8 papers
LoGRA: Scaling LLM Reinforcement Learning with Low-Rank Gradient Sketches
Shaokun Zhang, Yifan Zhang, Jian Hu +5
Reinforcement learning has greatly advanced the capabilities of large language models, but its memory demands remain a barrier to broader adoption. We introduce LoGRA, an approach…
Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents
Minki Kang, Ryo Hachiuma, Shaokun Zhang +8
Terminal agents act through stochastic model generations, yet the ability to generate a useful action does not ensure its reliable execution. A poor command (e.g., wrong package in…
OSWorld-Pro: Process-based Evaluation for Computer Use Agents
Zhilin Wang, Shaokun Zhang, Yifan Zhang +9
Evaluation of Computer-Use Agents (CUAs) is often limited to the final deliverables they create (at the end of hundreds of steps) and assessed with functional verifiers, as seen in…
Molt: A Scalable PyTorch-Native Training Framework for Agentic Reinforcement Learning
Jian Hu, Huiying Li, Hao Zhang +8
Agentic reinforcement learning requires rapid experimentation with agents and learning algorithms, yet large policies and long, multimodal trajectories demand substantial distribut…
ProCUA-SFT Technical Report
Jaehun Jung, Ximing Lu, Brandon Cui +11
Training computer-use agents (CUAs) -- models that interact with graphical desktops through screenshots and keyboard/mouse actions -- requires large-scale, diverse trajectory data…
Polar: Agentic RL on Any Harness at Scale
Binfeng Xu, Hao Zhang, Shaokun Zhang +9
Reinforcement learning for language agents increasingly depends on custom harnesses that manage long-running context, multi-turn tool use and multi-agent orchestration. However, po…
ProRL Agent: Rollout-as-a-Service for RL Training of Multi-Turn LLM Agents
Hao Zhang, Mingjie Liu, Shaokun Zhang +10
Multi-turn LLM agents are increasingly important for solving complex, interactive tasks, and reinforcement learning (RL) is a key ingredient for improving their long-horizon behavi…
Golden Goose: A Simple Trick to Synthesize Unlimited RLVR Tasks from Unverifiable Internet Text
Ximing Lu, David Acuna, Jaehun Jung +12
Reinforcement Learning with Verifiable Rewards (RLVR) has become a cornerstone for unlocking complex reasoning in Large Language Models (LLMs). Yet, scaling up RL is bottlenecked b…