ReGE 如何透過經驗合成提升智能體表現
ReGE: Stronger Agents through Experience Synthesis
ReGE 將智能體的執行記錄合成為指引,改善後續任務表現,無需更強教師模型或更新參數。四項基準測試中,相比無指引執行,ReGE 令 GPT 和 Claude 的準確率平均分別提升 9.9 和 8.8 個百分點。在 BrowseComp-Plus 的配對子集中,ReGE-Syn 的指引建構 token 數量分別減少 98.2% 和 95.5%。
Published on Oct 24
Authors:
,
,
,
,
Abstract
ReGE converts recorded agent execution experience into synthesized guidance to improve subsequent task attempts without requiring stronger teachers or parameter updates.
Repeated attempts at interactive tasks uncover facts, useful operations, and failures that can inform subsequent execution. We introduce ReGE, a framework for turning this recorded experience into guidance for a fresh attempt. Its main implementation, ReGE-Syn, uses a single coach call to jointly synthesize bounded execution records and available artifacts into self-contained guidance, connecting task-specific findings with proposed actions and relevant pitfalls. A student starts from the task's initial environment state and adapts this guidance to current observations. Source generation, synthesis, and student execution use the same base model, without parameter updates or a stronger teacher. Across four benchmarks, ReGE improves accuracy by an average of 9.9 percentage points for GPT and 8.8 percentage points for Claude compared with unguided execution. On matched BrowseComp-Plus subsets, ReGE-Syn reduces guidance-construction tokens by 98.2\% on GPT and 95.5\% on Claude relative to adapted AggAgent, with observed success. Combined construction and execution tokens decrease by 15.7\% and 35.4\%, respectively, counting input plus output and excluding source generation. BrowseComp-Plus ablations further associate access to process information with fewer student steps and show that joint synthesis produces substantially shorter guidance than Direct history or Independent summaries, with observed success. These results demonstrate how agents can use their own recorded experience to improve subsequent task completion, and suggest a low-cost way to consolidate the execution records of parallel agent systems such as agent swarms.
View arXiv page View PDF Project page GitHub Add to collection
Models citing this paper 0
No model linking this paper
Cite arxiv.org/abs/2610.33861 in a model README.md to link it from this page.
Datasets citing this paper 0
No dataset linking this paper
Cite arxiv.org/abs/2610.33861 in a dataset README.md to link it from this page.
Spaces citing this paper 0
No Space linking this paper
Cite arxiv.org/abs/2610.33861 in a Space README.md to link it from this page.
Collections including this paper 0
No Collection including this paper
Add this paper to a collection to link it from this page.
來源:HuggingFace Daily Papers(社區熱門論文) · huggingface.co