[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-84380-en":3,"doc-seo-84380-105":30,"detail-sidebar-cat-0-en-105":91},{"code":4,"msg":5,"data":6},0,"success",{"doc_id":7,"user_id":8,"nickname":9,"user_avatar":10,"doc_module":4,"category_id":11,"category_name":12,"doc_title":13,"doc_description":14,"doc_content":15,"file_id":16,"file_url":17,"file_type":18,"file_size":19,"view_count":20,"is_deleted":4,"is_public":21,"is_downloadable":21,"audit_status":21,"page_count":22,"language":23,"language_code":24,"site_id":25,"html_lang":24,"table_of_contents":26,"faqs":27,"seo_title":13,"seo_description":14,"update_tm":28,"read_time":29},84380,13056703020460,"Valentina","https://ap-avatar.wpscdn.com/avatar/be000253dac470eee5d?_k=1778207105932848923",8,"Research & Report","Improving Ad-hoc Search Effectiveness for Conversational Information Retrieval via Model Merging","Conversational information retrieval is difficult because it must exploit multi-turn conversation history, which introduces topic shifts and coreference that complicate interpreting users’ evolving information needs. Prior methods often rely on costly fine-tuning of ad-hoc retrievers for conversational data, leading to catastrophic forgetting and reduced ad-hoc performance. This work proposes a training-free model merging approach to combine separately adapted models into a single retriever. Experiments using Model Soup and Slerp show up to 15% higher NDCG@3 in zero-shot settings, improving ad-hoc effectiveness and generalization across task-specific datasets.","Improving Ad-hoc Search Effectiveness for Conversational Information Retrieval via Model Merging  \nAhmed Rayane Kebir  \nUniversity of Toulouse, IRIT Toulouse, France [ahmed-rayane.kebir@irit.fr](ahmed-rayane.kebir@irit.fr)  \nJose G. Moreno  \nUniversity of Toulouse, IRIT Toulouse, France [jose.moreno@irit.fr](jose.moreno@irit.fr)  \nLynda Tamine  \nUniversity of Toulouse, IRIT Toulouse, France [lynda.tamine@irit.fr](lynda.tamine@irit.fr)  \narXiv :2607 .08540v 1 [ cs .IR] 9 Jul 2026  \nAbstract  \nConversational information retrieval is challenging since it requires the consideration of the conversation history which potentially gives rise to topic shifts and coreference resolution across previous turns. To address these challenges, previous work mainly rely on traditional fine-tuning of ad-hoc retrievers on conversational datasets or extrapolates their generalizability through multi-tasking. However, this mainstream approach is costly—since it requires model re-training—and exhibits catastrophic forgetting, where the model loses its foundational ad-hoc retrieval performance. In this paper, we fill this gap by introducing model merging as a trainingfree strategy enabling the design of a single retrieval model that operates across both ad-hoc and conversational settings with no additional fine-tuning. We conduct experiments using linear and non-linear parameter-wise merging strategies—namely Model Soup and Slerp—on standard ad-hoc search and conversational retrieval datasets. Our results demonstrate that model merging significantly enhances the ad-hoc search capabilities of conversational retrievers while improving generalizability across task-specific datasets, achieving up to 15% higher NDCG@3 under zero-shot conditions.  \nCCS Concepts  \n• Information systems → Information Retrieval.  \nKeywords  \nConversational Search; Information Retrieval; Model Merging  \nACM Reference Format:  \nAhmed Rayane Kebir, Jose G. Moreno, and Lynda Tamine. 2026. Improving Ad-hoc Search Effectiveness for Conversational Information Retrieval via Model Merging. In Proceedings of the 49th International ACM SIGIR Conference on Research and Development in Information Retrieval (SIGIR ’26), July 20–24, 2026, Melbourne, VIC, Australia. ACM, New York, NY, USA, 6 pages. [https://doi.org/10.1145/3805712.3809939](https://doi.org/10.1145/3805712.3809939)  \n1 Introduction  \nConversational search is a well-established paradigm in which users engage in natural language, multi-turn interactions with a system to satisfy evolving information needs [17, 21] . A core component of conversational search is retrieving relevant documents by leveraging the conversational context in response to the user’s latest utterance. Compared to traditional ad-hoc retrieval, conversational information retrieval (CIR) introduces additional challenges. As interactions progress, the conversational history becomes longer  \nThis work is licensed under a Creative Commons Attribution 4 .0 International License. SIGIR ’26, Melbourne, VIC, Australia  \n© 2026 Copyright held by the owner/author(s) .  \nACM ISBN 979-8-4007-2599-9/2026/07  \n[https://doi.org/10.1145/3805712.3809939](https://doi.org/10.1145/3805712.3809939)  \nand noisier, often containing topic shifts, coreference, and linguistic ambiguities. Consequently, the latest user turn may be underspecified or context-dependent, increasing the difficulty of interpreting the information need and retrieving relevant documents [6] .  \nTo tackle these challenges, prior work has explored a variety of approaches for training dense retrievers tailored to conversational retrieval. Common methods typically fine-tune retrievers to operate over conversational sessions that include the current query, previous turns, system responses, and additional contextual signals [12, 14, 18] . Given the scarcity of large-scale conversational IR datasets, most approaches adapt pretrained ad-hoc retrievers—originally trained on large datasets such as MS MARCO—to conversation","cbCaiuXMvX9f5tdi","https://ap.wps.com/l/cbCaiuXMvX9f5tdi","pdf",992452,4,1,6,"English","en",105,"# Abstract\n# Introduction\n## Conversational search and its challenges\n## Prior approaches and limitations\n## Proposed model merging strategy","[{\"question\":\"What problem does the paper address in conversational information retrieval?\",\"answer\":\"It addresses the difficulty of retrieving relevant documents in multi-turn settings where conversation history adds topic shifts and coreference, making the latest query harder to interpret.\"},{\"question\":\"Why do conventional fine-tuning approaches underperform?\",\"answer\":\"They are costly because they require model re-training and they can cause catastrophic forgetting, degrading foundational ad-hoc retrieval capability.\"},{\"question\":\"How does the proposed method avoid additional training while improving performance?\",\"answer\":\"It uses training-free model merging to combine parameters from an ad-hoc retriever and its conversationally fine-tuned version, unifying robustness without further fine-tuning.\"}]",1784195207,15,{"code":4,"msg":31,"data":32},"ok",{"site_id":25,"language":24,"slug":33,"title":13,"keywords":34,"description":14,"schema_data":35,"social_meta":86,"head_meta":88,"extra_data":90,"updated_unix":28},"improving-ad-hoc-search-effectiveness-for-conversational-information-retrieval-via-model-merging","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,48,51],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":47},"https://docshare.wps.com/document/","Document",2,{"item":49,"name":12,"@type":43,"position":50},"https://docshare.wps.com/document/research-report/",3,{"item":52,"name":13,"@type":43,"position":20},"https://docshare.wps.com/document/improving-ad-hoc-search-effectiveness-for-conversational-information-retrieval-via-model-merging/84380/",{"url":52,"name":13,"@type":54,"author":55,"headline":13,"publisher":57,"fileFormat":60,"inLanguage":24,"description":14,"dateModified":61,"datePublished":62,"encodingFormat":60,"isAccessibleForFree":63,"interactionStatistic":64},"DigitalDocument",{"name":9,"@type":56},"Person",{"url":41,"name":58,"@type":59},"DocShare","Organization","application/pdf","2026-07-26","2026-07-16",true,{"@type":65,"interactionType":66,"userInteractionCount":20},"InteractionCounter",{"@type":67},"ViewAction",{"@type":69,"mainEntity":70},"FAQPage",[71,77,81],{"name":72,"@type":73,"acceptedAnswer":74},"What problem does the paper address in conversational information retrieval?","Question",{"text":75,"@type":76},"It addresses the difficulty of retrieving relevant documents in multi-turn settings where conversation history adds topic shifts and coreference, making the latest query harder to interpret.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"Why do conventional fine-tuning approaches underperform?",{"text":80,"@type":76},"They are costly because they require model re-training and they can cause catastrophic forgetting, degrading foundational ad-hoc retrieval capability.",{"name":82,"@type":73,"acceptedAnswer":83},"How does the proposed method avoid additional training while improving performance?",{"text":84,"@type":76},"It uses training-free model merging to combine parameters from an ad-hoc retriever and its conversationally fine-tuned version, unifying robustness without further fine-tuning.","https://schema.org",{"og:url":52,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":52},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,114,119,122,127,130,134],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":47,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":102,"show_sort_weight":103,"slug":104},"Exam",70,"exam",{"id":106,"doc_module":4,"doc_module_name":46,"category_name":107,"show_sort_weight":108,"slug":109},5,"Comic",60,"comic",{"id":22,"doc_module":4,"doc_module_name":46,"category_name":111,"show_sort_weight":112,"slug":113},"Technology",50,"technology",{"id":115,"doc_module":4,"doc_module_name":46,"category_name":116,"show_sort_weight":117,"slug":118},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":120,"slug":121},30,"research-report",{"id":123,"doc_module":4,"doc_module_name":46,"category_name":124,"show_sort_weight":125,"slug":126},9,"Religion & Spirituality",20,"religion-spirituality",{"id":125,"doc_module":4,"doc_module_name":46,"category_name":128,"show_sort_weight":125,"slug":129},"World Cup","world-cup",{"id":131,"doc_module":4,"doc_module_name":46,"category_name":132,"show_sort_weight":131,"slug":133},10,"Lifestyle","lifestyle",{"id":135,"doc_module":4,"doc_module_name":46,"category_name":136,"show_sort_weight":106,"slug":137},19,"General","general"]