This work develops an industrial workflow spanning prior and posterior optimization and establishes a systematic evaluation protocol covering both retrieval-side context and final answers, demonstrating the effectiveness of the framework and its industrial implementation.
Yun-Fei Zhong, Yin-Qiong Cai, Lixin Su et al.· 0 citations
Experiments show that Reflective On-policy Self-Distillation (ROSD), which distills transferable reasoning insights rather than complete reference trajectories, consistently outperforms standard OPSD and reinforcement learning baselines, better preserves reasoning diversity, stabilizes training, and mitigates late-stag...
Zi-Qi Zhao, Xin-Yu Ma, Liu Yang et al.· 16 citations
F2DR evaluates DeepSearch workflows across three dimensions: Content, Trajectory, and Answer, enabling comprehensive process-level assessment, and constructs DeepSearch RM-Bench, a dedicated benchmark for evaluating RMs in DeepSearch scenarios.
Bo-Jian Xiong, Wen-Tao Ding, Yu-Jing Lu et al.· 0 citations
Population-Guided Intent-Aware Rewriting (PGIR) is proposed, which captures dominant population intent via a Semantic Clustering Unit (SCU) and generates intent-aware rewrites through a Rewriting Unit (U), aligning queries with mainstream search goals without requiring user history.
Yuanzhao Guo, Shuai Zhang, Ang Li et al.· Annual International ACM SIG...· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
