Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
Cowork-Gym synthesizes interactive work environments at scale by employing a build-execute-revise loop that leverages execution evidence and human feedback to ensure consistency and repair generated environments. A stateful user simulator models realistic coworker dynamics through dialogue and workspace edits, enabling the training of large language models on complex, multi-tool interactions. Post-training on 64.9K synthesized environments, Qwen models show significant performance gains across agentic benchmarks, with Ours-27B achieving state-of-the-art results on specific benchmarks, surpassing larger models.
Synthesized interactive work environments enable LLMs to achieve state-of-the-art performance on complex agentic benchmarks, outperforming larger models with a 27B parameter model.
Cowork agents navigate dependencies across tools and files while adapting to changing user requirements and edits to shared workspaces. Scaling such environments requires maintaining consistency among task objectives, executable resources, user behavior, and success criteria. We introduce Cowork-Gym, a framework that synthesizes Cowork environments through a build-execute-revise loop that uses execution evidence and review feedback to validate and repair generated environments. To capture realistic Cowork dynamics, a stateful user simulator responds to the agent's actions through dialogue and direct edits to the shared workspace. We synthesize 64.9K environments across 14 application scenarios and use them to post-train 9B, 27B, and 35B-A3B Qwen models with supervised fine-tuning, reinforcement learning, and on-policy distillation. Averaged across six agentic benchmarks, these models outperform their corresponding base checkpoints by 8.7, 6.4, and 4.1 points, respectively. Ours-27B ranks second overall among the evaluated models and achieves the highest average scores on PawBench and ClawEval, outperforming models with over 100B total parameters on both benchmarks.
DAMO