外观
论文arXiv 2610.1319419 天前
JANUS：通过共演化世界模型增强长时程自主研究
原文链接：JANUS: Enhancing Long-Horizon Autonomous Research with Co-Evolving World Models（arXiv 2610.13194 · 发布日期：2026-10-09）
JANUS: Enhancing Long-Horizon Autonomous Research with Co-Evolving World Models
JANUS用共演化世界模型构建预测-验证-演化循环，缓解长时程自主研究执行反馈延迟，并在MLS-Bench-Lite和真实研究任务上验证。
核心要点
- 动机：生成-执行-反馈范式依赖反应式试错，昂贵执行导致反馈延迟。
- 方法：在演化图中由研究智能体提候选，世界模型预测价值、信息增益与探索效用。
- 机制：真实反馈被蒸馏为研究智能体技能、世界模型探索策略技能及协调协议。
- 结果：在MLS-Bench-Lite和真实研究任务上，JANUS显著优于现有最强基线。
验证范围与局限：验证范围为MLS-Bench-Lite和真实研究任务。
英文摘要 (Abstract)
Long-horizon autonomous research has shown promise in accelerating scientific progress through sustained cycles of hypothesis generation, implementation, and validation. However, this Generate-Execute-Feedback paradigm relies on reactive trial-and-error, where costly and time-consuming executions introduce severe feedback delays, throttling evolutionary progress under explosively expanding search space. To address this challenge, we propose JANUS, a framework that unifies future outcomes forecasting and past experience reflection through a Predict-Verify-Evolve loop driven by a co-evolving world model. Within an evolving graph, the research agent proposes candidates while world model scores them by forecasting value improvement, information gain, and exploration utility, then accordingly selects, refines, or prunes branches prior to execution. After receiving ground-truth feedback, JANUS distills experience into evolving skills: hierarchical skills for research agent, exploration-strategy skills for world model, and their coordination protocols. This mechanism enables continual co-evolution: the research agent sharpens candidate generation while the world model refines its predictive judgment over extended horizons. Extensive evaluations on MLS-Bench-Lite and real research tasks demonstrate that JANUS significantly outperforms state-of-the-art baselines, achieving both higher evolving efficiency and ceiling.
作者：Mengdi Liu, Chenghao Jia, Hong Chang, Mingjie He, Shiguang Shan, Xilin Chen
来源
本页事实对应的公开页面，以原文为准。