Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2601.22758 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912862643945472 |
|---|---|
| author | Qiu, Libin Gao, Zhirong Chen, Junfu Ye, Yuhang Huang, Weizhi Xue, Xiaobo Qiu, Wenkai Tang, Shuo |
| author_facet | Qiu, Libin Gao, Zhirong Chen, Junfu Ye, Yuhang Huang, Weizhi Xue, Xiaobo Qiu, Wenkai Tang, Shuo |
| contents | Large language model agents often fail to accumulate knowledge from experience, treating each task as an independent challenge. Recent methods extract experience as flattened textual knowledge, which cannot capture procedural logic of complex subtasks. They also lack maintenance mechanisms, causing repository degradation as experience accumulates. We introduce AutoRefine, a framework that extracts and maintains dual-form Experience Patterns from agent execution histories. For procedural subtasks, we extract specialized subagents with independent reasoning and memory. For static knowledge, we extract skill patterns as guidelines or code snippets. A continuous maintenance mechanism scores, prunes, and merges patterns to prevent repository degradation. Evaluated on ALFWorld, ScienceWorld, and TravelPlanner, AutoRefine achieves 98.4%, 70.4%, and 27.1% respectively, with 20-73% step reductions. On TravelPlanner, automatic extraction exceeds manually designed systems (27.1% vs 12.1%), demonstrating its ability to capture procedural coordination. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_22758 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | AutoRefine: From Trajectories to Reusable Expertise for Continual LLM Agent Refinement Qiu, Libin Gao, Zhirong Chen, Junfu Ye, Yuhang Huang, Weizhi Xue, Xiaobo Qiu, Wenkai Tang, Shuo Artificial Intelligence Large language model agents often fail to accumulate knowledge from experience, treating each task as an independent challenge. Recent methods extract experience as flattened textual knowledge, which cannot capture procedural logic of complex subtasks. They also lack maintenance mechanisms, causing repository degradation as experience accumulates. We introduce AutoRefine, a framework that extracts and maintains dual-form Experience Patterns from agent execution histories. For procedural subtasks, we extract specialized subagents with independent reasoning and memory. For static knowledge, we extract skill patterns as guidelines or code snippets. A continuous maintenance mechanism scores, prunes, and merges patterns to prevent repository degradation. Evaluated on ALFWorld, ScienceWorld, and TravelPlanner, AutoRefine achieves 98.4%, 70.4%, and 27.1% respectively, with 20-73% step reductions. On TravelPlanner, automatic extraction exceeds manually designed systems (27.1% vs 12.1%), demonstrating its ability to capture procedural coordination. |
| title | AutoRefine: From Trajectories to Reusable Expertise for Continual LLM Agent Refinement |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2601.22758 |