@inproceedings{sharma-etal-2026-test,
    title = "Test-Time Strategies for More Efficient and Accurate Agentic {RAG}",
    author = "Sharma, Abhinav  and
      Zhang, Brian  and
      Guntur, Deepti  and
      Zuo, Zhiyang  and
      Chaudhari, Shreyas  and
      Zhao, Wenlong  and
      Dernoncourt, Franck  and
      Mathur, Puneet  and
      Rossi, Ryan A.  and
      Lipka, Nedim",
    editor = "T.Y.S.S., Santosh  and
      Rodriguez, Juan Diego  and
      de Gibert, Ona",
    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, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-srw.41/",
    pages = "463--469",
    ISBN = "979-8-89176-393-7",
    abstract = "Retrieval-Augmented Generation (RAG) systems face challenges with complex, multi-hop questions, and iterative agentic frameworks such as Search-R1 (Jin et al., 2025) have been proposed to address these complexities. However, such approaches can introduce inefficiencies, including repetitive retrieval of previously processed information and challenges in contextualizing retrieved results effectively within the current generation prompt. Such issues can lead to unnecessary retrieval turns, suboptimal reasoning, inaccurate answers, and increased token consumption. In this paper, we investigate test-time modifications to Search-R1{'}s open-source Qwen2.5-7B pipeline to mitigate these identified shortcomings. Specifically, we explore the integration of two components and their combination: a contextualization module to better integrate relevant information from retrieved documents into reasoning, and a de-duplication module that replaces previously retrieved documents with the next most relevant ones. We evaluate our approaches using the HotpotQA (Yang et al., 2018) and the Natural Questions (Kwiatkowski et al., 2019) datasets, reporting the exact match (EM) score, an LLM-as-a-Judge assessment of answer correctness, and the average number of turns. Our best-performing variant (contextualization) achieves a 5.6{\%} increase in EM score and reduces the average number of turns by 10.5{\%} compared to the Search-R1 baseline. While contextualization itself introduces additional LLM calls, our results demonstrate improved answer accuracy and reduced retrieval load."
}