Knowledge Conflicts in Language Models

Follow this topic Ask about this topic

Knowledge conflicts arise when a model’s learned knowledge disagrees with information in a prompt, retrieved documents, tools, or stored conversation memory. Researchers test how models handle counterfactual and outdated facts, contradictory sources, and differences in framing or evidence order, using question-answering benchmarks, retrieval tasks, and multi-turn evaluations. Approaches include conflict-aware retrieval and decoding, confidence-based abstention, fine-tuning, and interventions that probe or steer internal representations. A central challenge is deciding when to follow context and when to resist it: stronger evidence-following can improve faithfulness but also make models more vulnerable to misleading, stale, or malicious information.

Latest papers

142 papers16 papers in the last four weeks, down 33% on the four weeks before. 0.1% of all new papers.

Language Bias under Conflicting Information in Multilingual LLMs
Apr 8, 2026Robert Östling, Murathan KurfalıLinguistic Bias in Language Models Language Model Bias Evaluation
Stockholm University Sweden · RISE Research Institutes of Sweden Sweden
ClawArena: Benchmarking AI Agents in Evolving Information Environments
Apr 5, 2026Haonian Ji, Kaiwen Xiong, Siwei Han +9AI Agent Evaluation AI Agent Benchmarks
UNC-Chapel Hill · University of California, Berkeley · University of California, Santa Cruz
When Do Hallucinations Arise? A Graph Perspective on the Evolution of Path Reuse and Path Compression
Apr 4, 2026Xinnan Dai, Kai Yang, Cheng Luo +3Hallucination in Language Models Knowledge Conflicts in Language Models
Michigan State University · University of Michigan · Independent Researcher
DeltaLogic: Minimal Premise Edits Reveal Belief-Revision Failures in Logical Reasoning Models
Apr 3, 2026Amit DhandaLLM Reasoning Belief Updating in LLMs
Amazon
In-Context Molecular Property Prediction with LLMs: A Blinding Study on Memorization and Knowledge Conflicts
Mar 26, 2026Matthias Busch, Marius Tacke, Sviatlana V. Lamaka +4LLM Evaluation Knowledge Conflicts in Language Models
Institute for Continuum and Material Mechanics, Technical University of Hamburg, Eißendorfer Straße, 21073 Hamburg, Germany · Institute of Material Systems Modeling, Helmholtz-Zentrum Hereon, Max-Planck-Straße, 21502 Geesthacht, Germany · Institute of Surface Science, Helmholtz-Zentrum Hereon, Max-Planck-Straße, 21502 Geesthacht, Germany +2
Tug-of-War within A Decade: Conflict Resolution in Vulnerability Analysis via Teacher-Guided Retrieval-Augmented Generations
Mar 25, 2026Ziyin Zhou, Jianyi Zhang, Xu ji +3Knowledge Conflicts in Language Models LLMs for Cybersecurity
Beijing Electronic Science and Technology Institute
Temporal Fact Conflicts in LLMs: Reproducibility Insights from Unifying DYNAMICQA and MULAN
Mar 16, 2026Ritajit Dey, Iadh Ounis, Graham McDonald +1Knowledge Conflicts in Language Models Temporal Reasoning in Language Models
University of Glasgow, Glasgow, UK · University of Strathclyde, Glasgow, UK
Learning to Forget: Sleep-Inspired Memory Consolidation for Resolving Proactive Interference in Large Language Models
Mar 15, 2026Ying XieKV-Cache Management Knowledge Conflicts in Language Models
Kennesaw State University
Truth as a Compression Artifact in Language Model Training
Mar 12, 2026Konstantin KrestnikovKnowledge Conflicts in Language Models LLM Honesty
Stop Listening to Me! How Multi-turn Conversations Can Degrade LLM Reliability
Mar 12, 2026Kevin H. Guo, Chao Yan, Avinash Baidya +5LLM Reliability Multi-Turn Dialogue Evaluation
Vanderbilt University · Vanderbilt University Medical Center · Intuit AI Research
Understanding the Interplay between LLMs' Utilisation of Parametric and Contextual Knowledge: A keynote at ECIR 2025
Mar 10, 2026Isabelle AugensteinKnowledge Conflicts in Language Models Factual Knowledge in Language Models
University of Copenhagen, Denmark
CC-VQA: Conflict- and Correlation-Aware Method for Mitigating Knowledge Conflict in Knowledge-Based Visual Question Answering
Feb 27, 2026Yuyang Hong, Jiaqi Gu, Yujin Lou +7Knowledge-Based VQA Knowledge Conflicts in Language Models
School of Artificial Intelligence, UCAS · MAIS, Institute of Automation · Alibaba Cloud Computing
When the Pure Reasoner Meets the Impossible Object: Analytic vs. Synthetic Fine-Tuning and the Suppression of Genesis in Language Models
Feb 26, 2026Amin AmouhadiLLM Fine-Tuning Knowledge Conflicts in Language Models
Institute for Artificial Intelligence, University of Georgia
ERA: Evidence-based Reliability Alignment for Honest Retrieval-Augmented Generation
Feb 24, 2026Sunguk Shin, Meeyoung Cha, Byung-Jun Lee +1RAG Robustness LLM Abstention
Korea University, Seoul, South Korea · MPI-SP, Bochum, Germany · KAIST, Daejeon, South Korea +1
Diagnosing Retrieval Bias Under Multiple In-Context Knowledge Updates in Large Language Models
Feb 18, 2026Boyu Qiao, Sean Guo, Xian Yang +4Knowledge Conflicts in Language Models Belief Updating in LLMs
Institute of Information Engineering, Chinese Academy of Sciences · School of Cyber Security, University of Chinese Academy of Sciences · Hong Kong University of Science and Technology +1
Diagnosing Knowledge Conflict in Multimodal Long-Chain Reasoning
Feb 16, 2026Jing Tang, Kun Wang, Haolang Lu +7Knowledge Conflicts in Language Models Multimodal Large Language Models
Huazhong University of Science and Technology · Nanyang Technological University · Beijing University of Posts and Telecommunications +2
To Copy or Not to Copy: Copying Is Easier to Induce Than Recall
Jan 17, 2026Mehrdad Farahani, Franziska Penzkofer, Richard JohanssonKnowledge Conflicts in Language Models LLM Interpretability
1Chalmers University of Technology · University of Gothenburg
Multi-Step Knowledge Interaction Analysis via Rank-2 Subspace Disentanglement
Nov 3, 2025Sekh Mainul Islam, Pepa Atanasova, Isabelle AugensteinDisentangled Representation Learning LLM Grounding
University of Copenhagen
The Chameleon Nature of LLMs: Quantifying Multi-Turn Stance Instability in Search-Enabled Language Models
Oct 19, 2025Shivam Ratnakar, Sanjay RaghavendraRetrieval-Augmented Generation LLM Evaluation
University of Southern California
When Tools Hurt LLM Reasoning: State-Dependent Belief Revision under External Evidence
Aug 21, 2025Yufeng Zhao, Junnan Liu, Hongwei Liu +4Belief Revision Knowledge Conflicts in Language Models
Shanghai Artificial Intelligence Laboratory · Department of Electronic Engineering, Tsinghua University · Beijing National Research Center for Information Science and Technology +2
Disentangling Reasoning Logic to Resolve Explicit Knowledge Conflicts
Aug 2, 2025Xianda Zheng, Zijian Huang, Meng-Fen Chiang +4Knowledge Conflicts in Language Models
School of Computer Science, University of Auckland · Department of Electronics and Electrical Engineering, National Yang Ming Chiao Tung University · School of Computing and Information Systems, Singapore Management University +1
ParamMute: Suppressing Knowledge-Critical FFNs for Faithful Retrieval-Augmented Generation
Feb 21, 2025Pengcheng Huang, Zhenghao Liu, Yukun Yan +8Retrieval-Augmented Generation Knowledge Conflicts in Language Models
School of Computer Science and Engineering, Northeastern University, China · Department of Computer Science and Technology, Institute for AI, Tsinghua University, China · Microsoft Research Asia, Beijing, China +1

Apr 8, 2026J/K move · Enter open · S save

Language Bias under Conflicting Information in Multilingual LLMs

Robert Östling, Murathan Kurfalı

Stockholm University Sweden · RISE Research Institutes of Sweden Sweden

Linguistic Bias in Language Models Language Model Bias Evaluation Multilingual Language Models Knowledge Conflicts in Language Models Multilingual Language Model Evaluation

Open paper AskPDF arXiv

Large Language Models (LLMs) have been shown to contain biases in the process of integrating conflicting information when answering questions. Here we ask whether such biases also exist with respect to which language is used for each conflicting piece of information. To answer this question, we extend the conflicting needles in a haystack paradigm to a multilingual setting and perform a comprehensive set of evaluations with naturalistic news domain data in five different languages, for a range of multilingual LLMs of different sizes. We find that all LLMs tested, including GPT-5.2, ignore the conflict and confidently assert only one of the possible answers in the large majority of cases. Furthermore, there is a consistent bias across models and prompting languages in which languages are preferred, with a general bias against Russian and, for the longest context lengths, in favor of Chinese. The language preferences are consistent between models trained inside and outside of mainland China, though somewhat stronger in the former category. There is also a general tendency among models to prioritize information that matches the language used for prompting. We hope to make users and developers of multilingual LLMs aware of this category of biases, to spur further research on their causes and possible mitigation.