ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909993334210560 |
|---|---|
| author | Yao, Yilun Huang, Shan Dai, Elsie Tan, Zhewen Duan, Zhenyu Jia, Shousheng Jiang, Yanbing Yang, Tong |
| author_facet | Yao, Yilun Huang, Shan Dai, Elsie Tan, Zhewen Duan, Zhenyu Jia, Shousheng Jiang, Yanbing Yang, Tong |
| contents | Large language models are increasingly deployed as research agents for deep search and long-horizon information seeking, yet their performance often degrades as interaction histories grow. This degradation, known as context rot, reflects a failure to maintain coherent and task-relevant internal states over extended reasoning horizons. Existing approaches primarily manage context through raw accumulation or passive summarization, treating it as a static artifact and allowing early errors or misplaced emphasis to persist. Motivated by this perspective, we propose ARC, which is the first framework to systematically formulate context management as an active, reflection-driven process that treats context as a dynamic internal reasoning state during execution. ARC operationalizes this view through reflection-driven monitoring and revision, allowing agents to actively reorganize their working context when misalignment or degradation is detected. Experiments on challenging long-horizon information-seeking benchmarks show that ARC consistently outperforms passive context compression methods, achieving up to an 11% absolute improvement in accuracy on BrowseComp-ZH with Qwen2.5-32B-Instruct. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2601_12030 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents Yao, Yilun Huang, Shan Dai, Elsie Tan, Zhewen Duan, Zhenyu Jia, Shousheng Jiang, Yanbing Yang, Tong Artificial Intelligence Large language models are increasingly deployed as research agents for deep search and long-horizon information seeking, yet their performance often degrades as interaction histories grow. This degradation, known as context rot, reflects a failure to maintain coherent and task-relevant internal states over extended reasoning horizons. Existing approaches primarily manage context through raw accumulation or passive summarization, treating it as a static artifact and allowing early errors or misplaced emphasis to persist. Motivated by this perspective, we propose ARC, which is the first framework to systematically formulate context management as an active, reflection-driven process that treats context as a dynamic internal reasoning state during execution. ARC operationalizes this view through reflection-driven monitoring and revision, allowing agents to actively reorganize their working context when misalignment or degradation is detected. Experiments on challenging long-horizon information-seeking benchmarks show that ARC consistently outperforms passive context compression methods, achieving up to an 11% absolute improvement in accuracy on BrowseComp-ZH with Qwen2.5-32B-Instruct. |
| title | ARC: Active and Reflection-driven Context Management for Long-Horizon Information Seeking Agents |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2601.12030 |