Paper Detail
Yunbo Tang, Chengyi Yang, Shiyu Liu, Zhishang Xiang, Zerui Chen, Qinggang Zhang, Jinsong Su
Agentic search enables LLMs to solve complex multi-hop questions through iterative reasoning and external search. Despite the effectiveness, these systems often suffer from a critical limitation in practice: agents fail to recognize their own knowledge boundaries, blindly triggering searches when internal knowledge suffices and failing to terminate search even when adequate evidence has been collected. The lack of self-awareness leads to severe over-search, incurring substantial inference latency and prohibitive computational cost. To this end, we propose SAAS, a novel RL framework designed to cultivate dynamic self-awareness that precisely regulates search behavior without compromising accuracy. SAAS introduces three key components: (i) a search boundary modeling mechanism, which identifies the search boundary under the evolving policy by contrasting search-disabled and search-enabled rollouts; (ii) a boundary-aware reward module, which translates this boundary awareness into trajectory-level penalties, suppressing unnecessary and redundant searches; and (iii) a stage-wise optimization strategy, which leverages a sequential curriculum to prioritize reasoning over search regularization, thereby avoiding reward hacking. Extensive experiments demonstrate that SAAS substantially reduces over-search, while maintaining accuracy. Our code is anonymously released at https://github.com/XMUDeepLIT/SAAS.
No structured notes yet. Add `summary_sections`, `why_relevant`, `claim_impact`, or `next_action` in `papers.jsonl` to enrich this view.
No ranking explanation is available yet.
No tags.
@misc{tang2026saas,
title = {SAAS: Self-Aware Reinforcement Learning for Over-Search Mitigation in Agentic Search},
author = {Yunbo Tang and Chengyi Yang and Shiyu Liu and Zhishang Xiang and Zerui Chen and Qinggang Zhang and Jinsong Su},
year = {2026},
abstract = {Agentic search enables LLMs to solve complex multi-hop questions through iterative reasoning and external search. Despite the effectiveness, these systems often suffer from a critical limitation in practice: agents fail to recognize their own knowledge boundaries, blindly triggering searches when internal knowledge suffices and failing to terminate search even when adequate evidence has been collected. The lack of self-awareness leads to severe over-search, incurring substantial inference latenc},
url = {https://huggingface.co/papers/2605.29796},
keywords = {agentic search, LLMs, multi-hop questions, iterative reasoning, external search, self-awareness, over-search, RL framework, search boundary modeling, boundary-aware reward module, stage-wise optimization, trajectory-level penalties, reward hacking, code available, huggingface daily},
eprint = {2605.29796},
archiveprefix = {arXiv},
}
{}