CP-Search: A Chain Progressive Search Training Framework Incentivizing the Cognitive Behaviors for Searching in LLMs
Zehua Wang, Shipeng Li, Buzhou Tang
Abstract
Retrieval-Augmented Generation (RAG) has been demonstrated to effectively mitigate the knowledge recency issue in Large Language Models (LLMs) while significantly reducing hallucinations. However, existing RAG methods exhibit insufficient capability in modeling reasoning paths for complex multi-hop reasoning tasks. While Reinforcement Learning (RL) has demonstrated success in enhancing model reasoning ability, Token-level RL frameworks exhibit inherent limitations in maintaining coherent reasoning trajectories. This approach remains susceptible to the compounding accumulation of contextual errors during the retrieval process, ultimately resulting in erroneous output generation. To address this challenge, we propose Chain Progressive Search (CP-Search), a novel two-stage training framework designed to enhance the model
BibTeX
@inproceedings{aaai2026_cpsearchachainpr,
title = {CP-Search: A Chain Progressive Search Training Framework Incentivizing the Cognitive Behaviors for Searching in LLMs},
author = {Zehua Wang and Shipeng Li and Buzhou Tang},
booktitle = {AAAI 2026},
year = {2026}
}