← Back ◬ AI & Machine Learning Apr 07, 2026

PRAISE: Prefix-Based Rollout Reuse in Agentic Search Training

arXiv AI Archived Apr 07, 2026 ✓ Full text saved

arXiv:2604.03675v1 Announce Type: new Abstract: In agentic search, large language models (LLMs) are trained to perform multi-turn retrieval and reasoning for complex tasks such as multi-hop question answering (QA). However, current search-based Reinforcement Learning (RL) methods suffer from two core limitations: expensive long-horizon rollouts are under-utilized during training, and supervision is typically available only at the final answer, resulting in severe reward sparsity. We present Pref

Full text archived locally

✦ AI Summary · Claude Sonnet

Computer Science > Artificial Intelligence [Submitted on 4 Apr 2026] PRAISE: Prefix-Based Rollout Reuse in Agentic Search Training Erhan Zhang, Yiqun Chen, Zechun Niu, Wei Yang, Xiaochi Wei, Yan Gao, Yi Wu, Yao Hu, Jiaxin Mao In agentic search, large language models (LLMs) are trained to perform multi-turn retrieval and reasoning for complex tasks such as multi-hop question answering (QA). However, current search-based Reinforcement Learning (RL) methods suffer from two core limitations: expensive long-horizon rollouts are under-utilized during training, and supervision is typically available only at the final answer, resulting in severe reward sparsity. We present Prefix-based Rollout reuse for Agentic search with Intermediate Step rEwards (PRAISE), a framework for improving both data efficiency and credit assignment in agentic search training. Given a complete search trajectory, PRAISE extracts prefix states at different search turns, elicits intermediate answers from them, and uses these prefixes both to construct additional training trajectories and to derive step-level rewards from performance differences across prefixes. Our method uses a single shared model for both search policy learning and prefix answer evaluation, enabling joint optimization without extra human annotations or a separate reward model. Experiments on multi-hop QA benchmarks show that PRAISE consistently improves performance over strong baselines. Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2604.03675 [cs.AI] (or arXiv:2604.03675v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2604.03675 Focus to learn more Submission history From: Erhan Zhang [view email] [v1] Sat, 4 Apr 2026 10:23:46 UTC (1,836 KB) Access Paper: HTML (experimental) view license Current browse context: cs.AI < prev | next > new | recent | 2026-04 Change to browse by: cs References & Citations NASA ADS Google Scholar Semantic Scholar Export BibTeX Citation Bookmark Bibliographic Tools Bibliographic and Citation Tools Bibliographic Explorer Toggle Bibliographic Explorer (What is the Explorer?) Connected Papers Toggle Connected Papers (What is Connected Papers?) Litmaps Toggle Litmaps (What is Litmaps?) scite.ai Toggle scite Smart Citations (What are Smart Citations?) Code, Data, Media Demos Related Papers About arXivLabs Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

💬 Team Notes