[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-detail-86323-en":3,"doc-seo-86323-105":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},86323,7971461741311,"Ophelia","https://ap-avatar.wpscdn.com/avatar/74000253aff267980c6?x-image-process=image/resize,m_fixed,w_180,h_180&k=1779345379180704826",8,"Research & Report","How Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation","Retrieval-Augmented Generation (RAG) increasingly grounds large language model (LLM) answers to reduce hallucinations, yet ideological bias embedded in retrieved sources can be transmitted, amplified, or suppressed. This study examines how RAG propagates ideological discourses into LLM-generated responses and how generation sampling temperature modulates the transfer strength. Using Lexical Multidimensional Analysis on 1,117 COVID-19 treatment articles, three ideological discourses are extracted and used as external knowledge for RAG evaluation across multiple LLMs.","arXiv :2607 . 1 1783v 1 [ cs .CL] 13 Jul 2026  \nPreprint – Not peer reviewed  \nARTICLE TYPE  \nHow Temperature Shapes Ideological Discourse in Retrieval-Augmented Generation?  \nE. Salari,*† H. Amamou,‡ J. V. De Souza,¶ S. Kshirsagar, § M. Nunes Delfino,|0 and A. Avila⊥†Elmira Salari, Wichita State University, Wichita, KS, USA  \n‡Hazem Amamou, Institut national de la recherche scientifique, Québec, Montréal, Canada ¶José Victor de Souza, Institut national de la recherche scientifique, Québec, Montréal, Canada |§Sh0MrautriKNsresasgaDrel,Wfinc,hSita Stateão PauloUCnaersoliyU, Wniviceitaity,, KSSão, UPSaAulo, Brazil  \n⊥Anderson Avila, Institut national de la recherche scientifique, Québec, Montréal, Canada  \n*Corresponding author. Email: [exsalari1@shockers.wichita.edu](exsalari1@shockers.wichita.edu)  \nAbstract  \nRetrieval-Augmented Generation (RAG) has been increasingly adopted to reduce hallucinations and strengthen the factual grounding of large language models (LLMs) . While robustness to errors in the retrieval process has been explored, the impact of ideological bias on LLM outputs has been overlooked. For instance, if the retrieved material contains ideological positions, the RAG may transmit, amplify, or suppress such ideological discourses in its outputs. In this study, we address this issue by examining the influence of the RAG framework, comprising ideological discourses, in LLM-generated answers. To this end, we applied Lexical Multidimensional Analysis (LMDA) on a corpus of 1,117 COVID-19 treatment articles, identifying three ideological discourses. This corpus is then used as the external knowledge source for the RAG. We assessed several LLMs by having the models answer ideological questions at different sampling temperatures. The generated texts were assessed semantically and lexically based on their similarities with ideological reference texts. Our findings show that the RAG framework is prone to transferring ideological discourses into LLM responses, with sampling temperature having a measurable impact on the strength of this transfer. Discoursive alignment between generated answers and the reference text is highest at moderate temperatures, where models balance stochasticity with retrieval grounding, and drops at low temperatures, indicating that overly deterministic sampling suppresses discourse transfer.  \nKeywords: Ideological discourse, retrieval-augmented generation, RAG, sampling temperature  \n1. Introduction  \nLarge Language Models (LLMs) have revolutionized the Artificial Intelligence paradigm, being increasingly used in domains such as healthcare, education, and finance, among others. Because such models may hallucinate providing incorrect answers for queries that require up-to-date or domainspecific knowledge, Retrieval-Augmented Generation (RAG) has been introduced as a solution to connect LLMs with external knowledge sources (Huang et al., 2025; Farquhar et al., 2024) . Notwithstanding, while RAG enhances factual analysis (Wallat et al., 2025), it also introduces new challenges, specifically in scenarios where the retrieved documents contain inaccurate information or ideological biases. Recent work, for instance, shows that LLMs’ responses contain not only factual cues but also ideological patterns (Holtzman et al., 2019; Lucy and Bamman, 2021; AlKhamissi et al., 2024) . When combined with RAG systems comprising ideological documents, it can potentially reinforce or shift ideological viewpoints, potentially influencing and steering public opinion. This is particularly concerning in sensitive contexts, such as healthcare, where changes in wording and emphasis can reshape how treatments, risks, or scientific debates are interpreted. Despite its  \n2 E. Salari et al.  \nrisks, limited attention has been paid to how RAG frameworks shape ideological discourse and how generation parameters can influence LLM outputs in such contexts. Sampling temperature, in particular, plays an important role in controlling ","cbCaipOKR3F2eeDb","https://ap.wps.com/l/cbCaipOKR3F2eeDb","pdf",1812432,2,1,23,"English","en",105,"# Introduction\n# Related Work","[{\"question\":\"What problem does the paper address in retrieval-augmented generation (RAG)?\",\"answer\":\"The paper addresses how ideological bias present in retrieved documents can shape the ideological content of LLM outputs, even when RAG is used to improve factual grounding and reduce hallucinations.\"},{\"question\":\"How are ideological discourses identified for use as external knowledge?\",\"answer\":\"Lexical Multidimensional Analysis (LMDA) is applied to a corpus of 1,117 COVID-19 treatment articles to identify three ideological discourses, which are then used as the RAG external knowledge source.\"},{\"question\":\"How does sampling temperature influence ideological transfer in generated answers?\",\"answer\":\"Sampling temperature measurably changes the strength of ideological transfer. Alignment is highest at moderate temperatures, while it drops at low temperatures because overly deterministic decoding suppresses discourse transfer.\"}]",1784210477,58,{"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},"how-temperature-shapes-ideological-discourse-in-retrieval-augmented-generation","",{"@graph":36,"@context":85},[37,53,68],{"@type":38,"itemListElement":39},"BreadcrumbList",[40,44,47,50],{"item":41,"name":42,"@type":43,"position":21},"https://docshare.wps.com","Home","ListItem",{"item":45,"name":46,"@type":43,"position":20},"https://docshare.wps.com/document/","Document",{"item":48,"name":12,"@type":43,"position":49},"https://docshare.wps.com/document/research-report/",3,{"item":51,"name":13,"@type":43,"position":52},"https://docshare.wps.com/document/how-temperature-shapes-ideological-discourse-in-retrieval-augmented-generation/86323/",4,{"url":51,"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","2026-07-21","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 retrieval-augmented generation (RAG)?","Question",{"text":75,"@type":76},"The paper addresses how ideological bias present in retrieved documents can shape the ideological content of LLM outputs, even when RAG is used to improve factual grounding and reduce hallucinations.","Answer",{"name":78,"@type":73,"acceptedAnswer":79},"How are ideological discourses identified for use as external knowledge?",{"text":80,"@type":76},"Lexical Multidimensional Analysis (LMDA) is applied to a corpus of 1,117 COVID-19 treatment articles to identify three ideological discourses, which are then used as the RAG external knowledge source.",{"name":82,"@type":73,"acceptedAnswer":83},"How does sampling temperature influence ideological transfer in generated answers?",{"text":84,"@type":76},"Sampling temperature measurably changes the strength of ideological transfer. Alignment is highest at moderate temperatures, while it drops at low temperatures because overly deterministic decoding suppresses discourse transfer.","https://schema.org",{"og:url":51,"og:type":87,"og:title":13,"og:site_name":58,"og:description":14},"article",{"robots":89,"canonical":51},"index,follow",{"doc_id":7,"site_id":25},{"code":4,"msg":5,"data":92},[93,97,101,105,110,115,120,123,128,131,135],{"id":21,"doc_module":4,"doc_module_name":46,"category_name":94,"show_sort_weight":95,"slug":96},"Story & Novel",90,"story-novel",{"id":20,"doc_module":4,"doc_module_name":46,"category_name":98,"show_sort_weight":99,"slug":100},"Literature",80,"literature",{"id":52,"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":111,"doc_module":4,"doc_module_name":46,"category_name":112,"show_sort_weight":113,"slug":114},6,"Technology",50,"technology",{"id":116,"doc_module":4,"doc_module_name":46,"category_name":117,"show_sort_weight":118,"slug":119},7,"Healthcare",40,"healthcare",{"id":11,"doc_module":4,"doc_module_name":46,"category_name":12,"show_sort_weight":121,"slug":122},30,"research-report",{"id":124,"doc_module":4,"doc_module_name":46,"category_name":125,"show_sort_weight":126,"slug":127},9,"Religion & Spirituality",20,"religion-spirituality",{"id":126,"doc_module":4,"doc_module_name":46,"category_name":129,"show_sort_weight":126,"slug":130},"World Cup","world-cup",{"id":132,"doc_module":4,"doc_module_name":46,"category_name":133,"show_sort_weight":132,"slug":134},10,"Lifestyle","lifestyle",{"id":136,"doc_module":4,"doc_module_name":46,"category_name":137,"show_sort_weight":106,"slug":138},19,"General","general"]