Search Wisely: Mitigating Sub-optimal Agentic Searches By Reducing Uncertainty

ACL ARR 2025 May Submission5733 Authors

Abstract: Agentic Retrieval-Augmented Generation (RAG) systems enhance Large Language Models (LLMs) by enabling dynamic, multi-step reasoning and information retrieval. However, these systems often exhibit sub-optimal search behaviors like over-search (retrieving redundant information) and under-search (failing to retrieve necessary information), which hinder efficiency and reliability. This work formally defines and quantifies these behaviors, revealing their prevalence across multiple QA datasets and agentic RAG systems (e.g., one model could have avoided searching in 27.7% of its search steps). Furthermore, we demonstrate a crucial link between these inefficiencies and the models’ uncertainty regarding their own knowledge boundaries, where response accuracy correlates with model’s uncertainty in its search decisions. To address this, we propose β-GRPO, a reinforcement learning-based training method that incorporates confidence threshold to reward high-certainty search decisions. Experiments on seven QA benchmarks show that β-GRPO enable a 3B model with better agentic RAG ability, outperforming other strong baselines with a 4% higher average exact match score.

Paper Type: Short

Research Area: NLP Applications

Research Area Keywords: NLP Applications, Retrieved Augmented Generation, LLM Agent, Language Modeling, Question Answering

Contribution Types: NLP engineering experiment, Data analysis

Languages Studied: English

Submission Number: 5733