@inproceedings{yuan-etal-2026-r3a,
    title = "{R}{\textthreesuperior}{A}: Reinforced Reasoning for Relevance Assessment for {RAG} in User-Generated Content Platforms",
    author = "Yuan, Xiaowei  and
      Jin, Lei  and
      Zhang, Haoxin  and
      Huang, Ziyang  and
      Gao, Yan  and
      Yiwu  and
      Hu, Yao  and
      Zhao, Jun  and
      Liu, Kang",
    editor = "Li, Yunyao  and
      Rehm, Georg  and
      Tu, Mei",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics ({ACL} 2026)",
    month = jul,
    year = "2026",
    address = "San Diego, California, USA",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-industry.10/",
    pages = "127--140",
    ISBN = "979-8-89176-394-4",
    abstract = "Retrieval-augmented generation (RAG) plays a critical role in user-generated content (UGC) platforms, but its effectiveness critically depends on accurate query{--}document relevance assessment. Despite recent advances in applying large language models (LLMs) to relevance modeling, UGC platforms present unique challenges: 1) ambiguous user intent due to sparse user feedback in RAG scenarios, and 2) asymmetric relevance, where relevance is driven by localized answer-bearing content rather than global query{--}document similarity. To address these issues, we propose the Reinforced Reasoning model for Relevance Assessment (R{\textthreesuperior}A), which decomposes relevance assessment into intent inference and evidence grounding. R{\textthreesuperior}A leverages auxiliary high-clicked documents to infer latent query intent, and extracts verbatim evidence fragments to ground relevance decisions, reducing noise sensitivity and improving asymmetric relevance modeling. Experimental results demonstrate that R{\textthreesuperior}A substantially outperforms strong baselines on offline benchmarks, while the distilled R{\textthreesuperior}A-1.5B model achieves significant gains in large-scale online A/B testing, effectively balancing performance and practical deployability."
}