QREF: Improving DeepSearch Agents by Aligning Search with Reasoning via Query Refinement and Evidence Filtering

Key points

• To mitigate these limitations, we propose QREF, a plug-and-play Query Refiner and Evidence Filter.1
• We initialize the Query Refiner with supervised fine-tuning and optimize it through reinforcement learning based on interactions with a real-world web search environment, reducing ineffective queries.1
• Comprehensive experiments demonstrate that our approach improves the performance of DeepSearch agents across four challenging benchmarks, achieving average relative Pass@1 improvements of 33.55% on WebSailor and 11.04% on TongyiDeepResearch without modifying architectures or retraining their reasoning policies.1
• Compared with the original DeepSearch backbone without QREF, the RL-trained refiner reduces the fraction of query batches that lack discriminative retrieval anchors by 95%.1

Sources (1)

• [1]QREF: Improving DeepSearch Agents by Aligning Search with Reasoning via Query Refinement and Evidence Filtering
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 15, 08:47 AM

  DeepSearch agents achieve strong performance on complex information-seeking tasks through iterative search and reasoning.

  To mitigate these limitations, we propose QREF, a plug-and-play Query Refiner and Evidence Filter.

Extractive summary: sentences quoted from the sources.