@inproceedings{burgess-etal-2026-papersearchqa,
    title = "{P}aper{S}earch{QA}: Learning to Search and Reason over Scientific Papers with {RLVR}",
    author = "Burgess, James  and
      Hansen, Jan N.  and
      Peng, Duo  and
      Zhang, Yuhui  and
      Lozano, Alejandro  and
      Sun, Min Woo  and
      Lundberg, Emma  and
      Yeung-Levy, Serena",
    editor = "Demberg, Vera  and
      Inui, Kentaro  and
      Marquez, Llu{\'i}s",
    booktitle = "Proceedings of the 19th Conference of the {E}uropean Chapter of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = mar,
    year = "2026",
    address = "Rabat, Morocco",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-eacl/2026.eacl-long.88/",
    pages = "1979--1997",
    ISBN = "979-8-89176-380-7",
    abstract = "Search agents are language models (LMs) that reason and search knowledge bases (or the web) to answer questions; recent methods supervise only the final answer accuracy using reinforcement learning with verifiable rewards (RLVR). Most RLVR search agents tackle general-domain QA, which limits their relevance to technical AI systems in science, engineering, and medicine. In this work we propose training agents to search and reason over scientific papers {--} this tests technical question-answering, it is directly relevant to real scientists, and the capabilities will be crucial to future AI Scientist systems. Concretely, we release a search corpus of 16 million biomedical paper abstracts and construct a challenging factoid QA dataset called PaperSearchQA with 60k samples answerable from the corpus, along with benchmarks. We train search agents in this environment to outperform non-RL retrieval baselines; we also perform further quantitative analysis and observe interesting agent behaviors like planning, reasoning, and self-verification. Our corpus, datasets, and benchmarks are usable with the popular Search-R1 codebase for RLVR training; they are available on Hugging Face. Finally, our data creation methods are scalable and easily extendable to other scientific domains."
}