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 Researchers test how models handle counterfactual and outdated facts, contradictory sources, and differences in framing or evidence order, using question-answering benchmarks, retrieval tasks, and multi-turn evaluations.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.

Coverage Under Contradiction: A Benchmark for Negation-Aware Retrieval
Oct 27, 2026Pasha Abrishamchian, Eugene Yang, Andrew Yates +2IR Evaluation Evidence Retrieval
Georgetown University Washington, DC, USA · Johns Hopkins University Baltimore, MD, USA
Do We Need an LLM to Know Why Sources Conflict? Decision Models for Knowledge-Conflict Typing in RAG
Oct 27, 2026Parv Khatri, Harshul Jain, Mehak Kapoor +1Knowledge Conflicts in Language Models Retrieval-Augmented Generation
Independent Researcher
Don't Pick a Winner: Knowledge Conflicts Need a Prior, Not a Rule
Oct 26, 2026Minh Pham Van, Cong TranKnowledge Conflicts in Language Models
Posts and Telecommunications Institute of Technology, Hanoi, Vietnam
Position Matters More Than Channel: Talking a Language Model Out of a Correct Answer
Oct 26, 2026Xinyi NiLLM Prompting Knowledge Conflicts in Language Models
Brandeis University
Accept or Repair: Preserving Passage-Supported Answers in Contrastive Decoding
Oct 21, 2026Jungseob Lee, Seungyoon Lee, Hyeonseok Moon +6Evidence-Grounded QA Language Model Decoding
Korea University · Sookmyung Women’s University · Yonsei University Mirae Campus +2
The Culture Within: Towards Mechanistic Understanding of Cultural Reasoning in LLMs
Oct 19, 2026Samee Arif, Furkan Danisman, Angana Borah +6Cultural Bias in Language Models Cultural Knowledge in Language Models
Shaping the Document, Not the Model: Optimized Suffixes for Context Following
Oct 19, 2026Gia-Bao Dinh Ho, Sekh Mainul Islam, Masahiro Kaneko +2Knowledge Conflicts in Language Models Automatic Prompt Optimization
Mohamed bin Zayed University of Artificial Intelligence · University of Copenhagen
When Retrieval Partially Conflicts with Skills: Understanding Procedural Hallucination in Language Models
Oct 19, 2026Guanqun Jiang, Pengfei Cao, Kang Liu +1Hallucination in Language Models Knowledge Conflicts in Language Models
The Key Laboratory of Cognition and Decision Intelligence for Complex Systems, Institute of Automation, Chinese Academy of Sciences, Beijing, China · Xi’an Jiaotong-Liverpool University, Suzhou, China · School of Artificial Intelligence, University of Chinese Academy of Sciences, Beijing, China
Text Dominance in Paralinguistic Conflict: Tracing Suppression Through Audio LLM Circuits
Oct 19, 2026Ankur Banerjee, Vishnu Guntakala, Danity Pike +3Audio-Language Models Audio-Language Model Evaluation
Independent Researcher · Brown University · University of Birmingham
Where and Why Search-Augmented LLMs Fail When Reasoning over Rapidly Evolving Knowledge
Oct 18, 2026Tianyi Hu, Bhavyajeet Singh, Niket Tandon +1Retrieval-Augmented Reasoning LLM Reasoning
Aarhus University · Microsoft Research · Technische Universität Darmstadt
MisleadQA: Diagnosing LLM Failures under Misleading Context
Oct 17, 2026Hui Wang, Bohao Chu, Qianli Wang +2LLM Evaluation Knowledge Conflicts in Language Models
University of Duisburg-Essen · Technische Universität Berlin · Tsinghua University
Retrieved but Not Read: Understanding Why Search-Augmented LLMs Fail to Use Available Evidence
Oct 15, 2026Owen Walters, Lihu Chen, Francesca ToniRetrieval-Augmented Generation LLM Reliability
Department of Computing, Imperial College London, UK
Which Knowledge Should LLMs Trust? Uncovering a Generalizable Steering Direction Across Knowledge Conflicts
Oct 14, 2026Yibin Lou, Yi Hu, Jianhui Chen +3Knowledge Conflicts in Language Models LLM Grounding
State Key Laboratory of Multimedia Information Processing, Peking University · School of Computer Science, Peking University · Southern University of Science and Technology +2
Knowledge Interaction in Hypernetwork-based Continual Learning: Conflict and Composition
Oct 11, 2026Jiaqi Zheng, Yi Hu, Yadong Mu +1Continual Learning for LLMs Knowledge Conflicts in Language Models
Peking University · Norizon AI
Accurate but Not Humble: Evaluating Epistemic Humility in LLM Agents under Knowledge Conflict
Oct 8, 2026Kaiser Sun, Bernal Jimenez Gutierrez, Hongjun Liu +4LLM Agent Evaluation Epistemic Uncertainty
Johns Hopkins University · New York University
RAG-Stress: Probing the Limits of Evidence Reliance in Retrieval-Augmented Generation
Oct 8, 2026Shunyuan Zhou, Hao Chen, Tianyu Wang +3Retrieval-Augmented Generation RAG Evaluation
Beijing University of Posts and Telecommunications · Humanlaya Data · Beijing Key Laboratory of Key Technologies for AI+ Domain Applications +1
The Attribution Blind Spot: Layerwise Trajectory Diagnostics for Source Reliance in Retrieval-Augmented Language Models
Oct 7, 2026Zhe Yu, Wenpeng Xing, Yunzhao Wei +4LLM Interpretability Retrieval-Augmented Generation
Zhejiang University · Binjiang Institute of Zhejiang University · East China Normal University +3
Where's Waldo? Query-language Preference under Cross-lingual Knowledge Disparities
Sep 30, 2026Dayeon Ki, Ruochen Zhang, Silviu Cucerzan +2Multilingual Language Model Evaluation Knowledge Conflicts in Language Models
University of Maryland · Microsoft
Can AI Scientists Change Their Minds? Prior-Evidence Conflict in Synthetic Universes
Sep 29, 2026Kargi ChauhanKnowledge Conflicts in Language Models Mechanistic Interpretability
University of California, Santa Cruz
Epistemic Policy Divergence in Multi-Turn LLM Contamination: A Protocol-Gradient Investigation
Sep 28, 2026Fahrell Giovanny, Geby Bayuningtyas, Sahrul Mukharom +1Knowledge Conflicts in Language Models Adversarial Attacks on LLMs
Independent Researcher, Tokyo, Japan · Independent Researcher, Jakarta, Indonesia · Department of Informatics, Institut Teknologi Sumatera, Lampung, Indonesia
Positions Are Not Facts: The Mismatch Between KV Caches and Memory
Sep 27, 2026Changhai Zhou, Yuhua Zhou, Shiyang Zhang +5Knowledge Editing Knowledge Conflicts in Language Models
Fudan University · Baidu · Zhejiang University +2
Reset Is Not Recovery: Evaluating Recoverability from False Conversational Context via Sycophancy Hysteresis
Sep 27, 2026Adi ShnaidmanLLM Sycophancy Knowledge Conflicts in Language Models
University of Pennsylvania · Work conducted while affiliated with DeepKeep.
Stale-Document Poisoning: When Outdated Retrieval Overrides Correct Model Answers
Sep 25, 2026Md Shamim Ahmed, Lukas Galke Poech, Richard RöttgerRAG Poisoning Attacks RAG Robustness
University of Southern Denmark
Words Speak Louder Than Order: A Behavioral Evaluation of Gemma 4
Sep 25, 2026Amanda FitchFraming Effects in Language Models Language Model Bias Evaluation
Google
Parameters vs. Context: TRACE Fine-Tuning for Robust Retrieval-Augmented Generation
Sep 24, 2026Zhengchen Huang, Yundong Sun, Minrui Song +4RAG Robustness Retrieval-Augmented Generation
School of Information Science & Engineering, LinYi University, China · School of Computer Science and Technology, Harbin Institute of Technology, China
Persuaded, Not Informed: Incentive-Misaligned Witnesses Defeat In-Context Grounding
Sep 23, 2026Rahul BalakaviKnowledge Conflicts in Language Models Agent Evaluation
AmpUp Research
Rewired or Gated? How Instruction Tuning Shapes Knowledge-Conflict Circuits in LLMs
Sep 22, 2026Shubham Santosh Pandere, Gautam Ranka, Ritika Varshney +3Knowledge Conflicts in Language Models LLM Interpretability
IvLabs, VNIT
Memory vs. Context? Influential Factors of Factual Recall in Language Models
Sep 21, 2026Guilhem Fouilhé, Nicholas Asher, Philippe MullerPrompt Sensitivity Memorization in Language Models
IRIT ANITI · IRIT CNRS · IRIT Université de Toulouse
An Interpretable Memory Decision Controller for LLM Agents Based on Three-Signal Complementarity: Decoupling Confidence and Consistency
Sep 18, 2026Yiming Zhang, Jinghong Zhang, Haoran Zhao +2Memory-Augmented Language Models LLM Hallucination Mitigation
School of Computer Science and Engineering, Tianjin University of Technology Tianjin, China
The Temporal Tug-of-War: Visualizing and Detecting RAG Conflicts in Diffusion Models via Trajectory Variance
Sep 16, 2026Sravan Karthick T, Pranav Darshan, Pranav A +2Knowledge Conflicts in Language Models Retrieval-Augmented Generation
Department of Computer Science and Engineering, R.V. College of Engineering, India · CAIRO, Technical University of Applied Sciences Würzburg-Schweinfurt, Germany

Oct 27, 2026J/K move · Enter open · S save

Coverage Under Contradiction: A Benchmark for Negation-Aware Retrieval

Pasha Abrishamchian, Eugene Yang, Andrew Yates, Nazli Goharian+1

Georgetown University Washington, DC, USA · Johns Hopkins University Baltimore, MD, USA

IR Evaluation Evidence Retrieval Knowledge Conflicts in Language Models

Open paper AskPDF arXiv

Coverage-evaluation benchmarks quantify whether retrieved evidence covers the different pieces of information needed to satisfy an information need. Existing benchmarks mostly consider factual claims that may already be known to pretrained retrieval components, leaving open whether retrievers find corpus evidence when it contradicts those priors. We study how retrievers behave when their parametric knowledge and the information in a corpus disagree. To do so, we develop a process to augment a nugget-based coverage benchmark with controlled contradictions. For each nugget, we create a negated counterpart and edit its supporting documents to support that negation while preserving the context. Applying this process to CoverageBench, a benchmark measuring information coverage, we create CoverageBench-Negate, spanning six datasets and roughly 63,000 original-edited evidence pairs (one per nugget-document link). We attach rich construction metadata to every pair and additionally provide out-of-family LLM judgments over all pairs and a human audit of a sample. In baseline experiments, BM25 separates original from edited near-twin evidence substantially better than dense retrievers (Qwen3 and Jina), which stay near chance. We release the benchmark on Hugging Face Datasets (nuggets, documents, pairs) together with the construction materials, including the prompts used to create it.