@inproceedings{xie-etal-2026-searching,
    title = "Over-Searching in Retrieval-Augmented Large Language Models",
    author = "Xie, Roy  and
      Gopinath, Deepak  and
      Qiu, David  and
      Lin, Dong  and
      Sun, Haitian  and
      Potdar, Saloni  and
      Dhingra, Bhuwan",
    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.361/",
    pages = "7714--7739",
    ISBN = "979-8-89176-380-7",
    abstract = "Search-augmented large language models (LLMs) excel at knowledge-intensive tasks by integrating external retrieval. However, they often over-search {--} unnecessarily invoking search tool even when it does not improve response quality, which leads to computational inefficiency and hallucinations by incorporating irrelevant context. In this work, we conduct a systematic evaluation of over-searching across multiple dimensions, including query types, model categories, retrieval conditions, and multi-turn conversations. Our findings show: (i) search generally improves answer accuracy on answerable queries but harms abstention on unanswerable ones; (ii) over-searching is more pronounced in complex reasoning models and deep research systems, is exacerbated by noisy retrieval, and compounds across turns in multi-turn conversations; and (iii) the composition of retrieved evidence is crucial, as the presence of negative evidence improves abstention. To quantify over-searching, we introduce Tokens Per Correctness (TPC), an evaluation metric that captures the performance-cost trade-off for search-augmented LLMs. Lastly, we investigate mitigation approaches at both the query and retrieval levels and release the OverSearchQA benchmark to foster continued research into efficient search-augmented LLMs."
}