[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"doc-seo-273472-105":3,"detail-sidebar-cat-0-en-105":81,"doc-detail-273472-en":131},{"code":4,"msg":5,"data":6},0,"ok",{"site_id":7,"language":8,"slug":9,"title":10,"keywords":11,"description":12,"schema_data":13,"social_meta":74,"head_meta":76,"extra_data":78,"updated_unix":80},105,"en","when-context-misleads-intent-guided-decoding-for-robust-retrieval-augmented-generation","When Context Misleads - Intent-Guided Decoding for Robust Retrieval-Augmented Generation","","Retrieval-augmented generation (RAG) grounds large language model outputs in external evidence, yet it creates a source-trust challenge: retrieved context can be useful, irrelevant, or actively misleading. Many RAG systems apply a fixed trust policy, over-trusting incorrect context or underusing context when users want strict context-following behavior. This work proposes Intent-Guided Decoding (IGD), which arbitrates between retrieved context and parametric memory using answer-level filtering and token-level correction to improve factual recovery while preserving intent-aligned faithfulness.",{"@graph":14,"@context":73},[15,34,56],{"@type":16,"itemListElement":17},"BreadcrumbList",[18,23,27,31],{"item":19,"name":20,"@type":21,"position":22},"https://docshare.wps.com","Home","ListItem",1,{"item":24,"name":25,"@type":21,"position":26},"https://docshare.wps.com/document/","Document",2,{"item":28,"name":29,"@type":21,"position":30},"https://docshare.wps.com/document/research-report/","Research & Report",3,{"item":32,"name":10,"@type":21,"position":33},"https://docshare.wps.com/document/when-context-misleads-intent-guided-decoding-for-robust-retrieval-augmented-generation/273472/",4,{"url":32,"name":10,"@type":35,"image":36,"author":41,"headline":10,"publisher":44,"fileFormat":47,"inLanguage":8,"description":12,"dateModified":48,"datePublished":49,"encodingFormat":47,"isAccessibleForFree":50,"interactionStatistic":51},"DigitalDocument",{"url":37,"@type":38,"width":39,"height":40},"https://docshare.wps.com/thumbnails/when-context-misleads-intent-guided-decoding-for-robust-retrieval-augmented-generation/273472.png","ImageObject",300,407,{"name":42,"@type":43},"Emma Wilson","Person",{"url":19,"name":45,"@type":46},"DocShare","Organization","application/pdf","2026-10-25","2026-09-15",true,{"@type":52,"interactionType":53,"userInteractionCount":55},"InteractionCounter",{"@type":54},"ViewAction",15,{"@type":57,"mainEntity":58},"FAQPage",[59,65,69],{"name":60,"@type":61,"acceptedAnswer":62},"What core issue does the document identify in retrieval-augmented generation (RAG)?","Question",{"text":63,"@type":64},"RAG can retrieve context that is misleading or contradictory, creating a source-trust problem. This leads to outputs that may be unsupported by the context or even conflict with stable world knowledge.","Answer",{"name":66,"@type":61,"acceptedAnswer":67},"What is Intent-Guided Decoding (IGD) and how does it work?",{"text":68,"@type":64},"IGD arbitrates between retrieved context and parametric memory according to user intent. It uses answer-level filtering to route high-confidence cases and token-level correction to adjust decoding when intervention is needed.",{"name":70,"@type":61,"acceptedAnswer":71},"How does IGD balance factuality with context-following behavior?",{"text":72,"@type":64},"IGD improves factual recovery on factual-conflict benchmarks while preserving or improving strict context-following behavior. The approach tailors trust in retrieved context rather than applying a fixed policy.","https://schema.org",{"og:url":32,"og:type":75,"og:title":10,"og:site_name":45,"og:description":12},"article",{"robots":77,"canonical":32},"index,follow",{"doc_id":79,"site_id":7},273472,1789471582,{"code":4,"msg":82,"data":83},"success",[84,88,92,96,101,106,111,115,120,123,127],{"id":22,"doc_module":4,"doc_module_name":25,"category_name":85,"show_sort_weight":86,"slug":87},"Story & Novel",90,"story-novel",{"id":26,"doc_module":4,"doc_module_name":25,"category_name":89,"show_sort_weight":90,"slug":91},"Literature",80,"literature",{"id":33,"doc_module":4,"doc_module_name":25,"category_name":93,"show_sort_weight":94,"slug":95},"Exam",70,"exam",{"id":97,"doc_module":4,"doc_module_name":25,"category_name":98,"show_sort_weight":99,"slug":100},5,"Comic",60,"comic",{"id":102,"doc_module":4,"doc_module_name":25,"category_name":103,"show_sort_weight":104,"slug":105},6,"Technology",50,"technology",{"id":107,"doc_module":4,"doc_module_name":25,"category_name":108,"show_sort_weight":109,"slug":110},7,"Healthcare",40,"healthcare",{"id":112,"doc_module":4,"doc_module_name":25,"category_name":29,"show_sort_weight":113,"slug":114},8,30,"research-report",{"id":116,"doc_module":4,"doc_module_name":25,"category_name":117,"show_sort_weight":118,"slug":119},9,"Religion & Spirituality",20,"religion-spirituality",{"id":118,"doc_module":4,"doc_module_name":25,"category_name":121,"show_sort_weight":118,"slug":122},"World Cup","world-cup",{"id":124,"doc_module":4,"doc_module_name":25,"category_name":125,"show_sort_weight":124,"slug":126},10,"Lifestyle","lifestyle",{"id":128,"doc_module":4,"doc_module_name":25,"category_name":129,"show_sort_weight":97,"slug":130},19,"General","general",{"code":4,"msg":82,"data":132},{"doc_id":79,"user_id":133,"nickname":42,"user_avatar":134,"doc_module":4,"category_id":112,"category_name":29,"doc_title":10,"doc_description":12,"doc_content":135,"file_id":136,"file_url":137,"file_type":138,"file_size":139,"view_count":55,"is_deleted":4,"is_public":22,"is_downloadable":22,"audit_status":22,"page_count":140,"language":141,"language_code":8,"site_id":7,"html_lang":8,"table_of_contents":142,"faqs":143,"seo_title":144,"seo_description":12,"update_tm":80,"read_time":145,"preview_page_count":4},3848291630094,"https://eur-avatar.wpscdn.com/davatar_085a072bc5b1113ac321206ff7593b45","When Context Misleads: Intent-Guided Decoding for Robust Retrieval-Augmented Generation  \nHaolin Jin, Pengyue Yang, Huaming Chen  \nSchool of Electrical and Computer Engineering  \nThe University of Sydney  \nSydney, NSW, Australia  \n{haolin.jin, pengyue.yang, [huaming.chen}@sydney.edu.au](huaming.chen}@sydney.edu.au)  \narXiv :2608 . 16515v1 [ cs .CL] 17 Aug 2026  \nAbstract—Retrieval-augmented generation (RAG) improves large language models by grounding generation in external evidence, but it also introduces a source trust problem: retrieved context may be useful, irrelevant, or even misleading. Existing RAG systems often apply a fixed trust policy toward retrieved evidence, which can either over-trust incorrect context or underuse context when the user explicitly asks for context-following behavior. Therefore, we propose Intent-Guided Decoding (IGD), a framework that arbitrates between retrieved context and parametric memory according to user intent. IGD uses answerlevel filtering and token-level correction to steer the final decoding trajectory between retrieved context and parametric memory. We evaluate IGD on three faithful QA benchmarks and three factualconflict benchmarks across five LLMs, IGD substantially improves factual recovery, achieving gains of up to 65.4 percentage points on factual-conflict benchmarks over Direct RAG, while preserving or improving strict context-following behavior, this findings highlight the importance of balancing factuality and faithfulness in RAG.  \nIndex Terms—Retrieval-Augmented Generation, Large Language Models, Factuality, Faithfulness  \nI. INTRODUCTION  \nRetrieval-augmented generation (RAG) has become a widely adopted paradigm for connecting large language models (LLMs) with external knowledge. By retrieving relevant evidence at inference time, retrieval-augmented generation extends the static parametric knowledge of LLMs with updatable nonparametric memory [1] . This design improves performance on knowledge-intensive tasks such as open-domain question answering, where models require access to facts beyond their internal parametric knowledge [2]–[4] .  \nHowever, the effectiveness of RAG depends not only on whether useful evidence is retrieved, but also on whether the model can ground the generation in the retrieved evidence appropriately. In practice, RAG systems may still generate statements that are unsupported by, or even contradictory to, the provided context. RAGTruth systematically annotates nearly 18K RAG responses, and finds that non-factual statements remain common under standard RAG pipelines [5] . On the other hand, retrieval introduces a source-trust problem. Standard RAG pipelines often implicitly treat retrieved context as authoritative, which content may be false, adversarially corrupted, or inconsistent with stable world knowledge [6] . FaithEval formalizes this issue as contextual faithfulness evaluation, and constructs 4.9K high-quality examples across unanswerable, inconsistent,  \nand counterfactual settings, showing that even strong contemporary LLMs often fail to maintain an appropriate degree of faithfulness to given context [7] . More broadly, situated faithfulness argues that models should dynamically calibrate their trust in external context against internal knowledge, rather than blindly following either source [8] . These findings suggest that a central challenge in RAG is not only how to retrieve useful evidence, but also how to decide whether the retrieved evidence should dominate generation [9], [10] .  \nThis issue exposes a fundamental trade-off between faithfulness and factuality in RAG generation. In some settings, the user explicitly expects the model to follow the provided context, for example when answering questions about a document, policy, or passage [11] . In other settings, the user expects the model to use retrieved context only as auxiliary evidence and remain skeptical when that context is misleading or conflicts with world knowledge [9], [10] .","cbCaijnf9lwgs8eC","https://ap.wps.com/l/cbCaijnf9lwgs8eC","pdf",557472,11,"English","# Introduction\n## Problem: faithfulness vs factuality in RAG\n## Proposed solution: Intent-Guided Decoding (IGD)","[{\"question\":\"What core issue does the document identify in retrieval-augmented generation (RAG)?\",\"answer\":\"RAG can retrieve context that is misleading or contradictory, creating a source-trust problem. This leads to outputs that may be unsupported by the context or even conflict with stable world knowledge.\"},{\"question\":\"What is Intent-Guided Decoding (IGD) and how does it work?\",\"answer\":\"IGD arbitrates between retrieved context and parametric memory according to user intent. It uses answer-level filtering to route high-confidence cases and token-level correction to adjust decoding when intervention is needed.\"},{\"question\":\"How does IGD balance factuality with context-following behavior?\",\"answer\":\"IGD improves factual recovery on factual-conflict benchmarks while preserving or improving strict context-following behavior. The approach tailors trust in retrieved context rather than applying a fixed policy.\"}]","When Context Misleads - Intent-Guided Decoding for Robust Retrieval-Augmented Generation | PDF",28]