Image de couverture de livre

Believe It or Not: How Deeply do LLMs Believe Implanted Facts?

Fuente: arXiv

Enregistré dans:

Détails bibliographiques
Auteurs principaux:	Slocum, Stewart, Minder, Julian, Dumas, Clément, Sleight, Henry, Greenblatt, Ryan, Marks, Samuel, Wang, Rowan
Format:	Preprint
Publié:	2025
Sujets:	Computation and Language Artificial Intelligence
Accès en ligne:	Acceder al recurso
Tags:	Ajouter un tag Pas de tags, Soyez le premier à ajouter un tag!

Documents similaires

Narrow Finetuning Leaves Clearly Readable Traces in Activation Differences
par: Minder, Julian, et autres
Publié: (2025)

To Believe or Not to Believe Your LLM
par: Yadkori, Yasin Abbasi, et autres
Publié: (2024)

Overcoming Sparsity Artifacts in Crosscoders to Interpret Chat-Tuning
par: Minder, Julian, et autres
Publié: (2025)

Activation Oracles: Training and Evaluating LLMs as General-Purpose Activation Explainers
par: Karvonen, Adam, et autres
Publié: (2025)

Seeing Isn't Believing: Mitigating Belief Inertia via Active Intervention in Embodied Agents
par: Wang, Hanlin, et autres
Publié: (2026)

How Far Are LLMs from Believable AI? A Benchmark for Evaluating the Believability of Human Behavior Simulation
par: Xiao, Yang, et autres
Publié: (2023)

Don't Believe Everything You Read: Enhancing Summarization Interpretability through Automatic Identification of Hallucinations in Large Language Models
par: Vakharia, Priyesh, et autres
Publié: (2023)

How LLMs Fail to Support Fact-Checking
par: Proma, Adiba Mahbub, et autres
Publié: (2025)

All Code, No Thought: Current Language Models Struggle to Reason in Ciphered Language
par: Guo, Shiyuan, et autres
Publié: (2025)

Controllable Context Sensitivity and the Knob Behind It
par: Minder, Julian, et autres
Publié: (2024)

Seeing is Believing: Mitigating Hallucination in Large Vision-Language Models via CLIP-Guided Decoding
par: Deng, Ailin, et autres
Publié: (2024)

Stress-Testing Model Specs Reveals Character Differences among Language Models
par: Zhang, Jifan, et autres
Publié: (2025)

Is Seeing Believing? Evaluating Human Sensitivity to Synthetic Video
par: Wegmann, David, et autres
Publié: (2026)

Achieving >97% on GSM8K: Deeply Understanding the Problems Makes LLMs Better Solvers for Math Word Problems
par: Zhong, Qihuang, et autres
Publié: (2024)

Unsupervised Elicitation of Language Models
par: Wen, Jiaxin, et autres
Publié: (2025)

FASTTRACK: Fast and Accurate Fact Tracing for LLMs
par: Chen, Si, et autres
Publié: (2024)

Separating Tongue from Thought: Activation Patching Reveals Language-Agnostic Concept Representations in Transformers
par: Dumas, Clément, et autres
Publié: (2024)

Can LLMs Automate Fact-Checking Article Writing?
par: Sahnan, Dhruv, et autres
Publié: (2025)

InFi-Check: Interpretable and Fine-Grained Fact-Checking of LLMs
par: Bai, Yuzhuo, et autres
Publié: (2026)

Fact or Fiction? Can LLMs be Reliable Annotators for Political Truths?
par: Chatrath, Veronica, et autres
Publié: (2024)

MiniCheck: Efficient Fact-Checking of LLMs on Grounding Documents
par: Tang, Liyan, et autres
Publié: (2024)

Subjective Experience in AI Systems: What Do AI Researchers and the Public Believe?
par: Dreksler, Noemi, et autres
Publié: (2025)

Robustly Improving LLM Fairness in Realistic Settings via Interpretability
par: Karvonen, Adam, et autres
Publié: (2025)

Believing is Seeing: Unobserved Object Detection using Generative Models
par: Bhattacharjee, Subhransu S., et autres
Publié: (2024)

FactSelfCheck: Fact-Level Black-Box Hallucination Detection for LLMs
par: Sawczyn, Albert, et autres
Publié: (2025)

When Facts Change: Probing LLMs on Evolving Knowledge with evolveQA
par: Nakshatri, Nishanth Sridhar, et autres
Publié: (2025)

Seeing Is No Longer Believing: Frontier Image Generation Models, Synthetic Visual Evidence, and Real-World Risk
par: Wu, Shuai, et autres
Publié: (2026)

Seeing but Not Believing: Probing the Disconnect Between Visual Attention and Answer Correctness in VLMs
par: Liu, Zhining, et autres
Publié: (2025)

CANDY: Benchmarking LLMs' Limitations and Assistive Potential in Chinese Misinformation Fact-Checking
par: Guo, Ruiling, et autres
Publié: (2025)

Censored LLMs as a Natural Testbed for Secret Knowledge Elicitation
par: Casademunt, Helena, et autres
Publié: (2026)

Liars' Bench: Evaluating Lie Detectors for Language Models
par: Kretschmar, Kieron, et autres
Publié: (2025)

Facts Do Care About Your Language: Assessing Answer Quality of Multilingual LLMs
par: Kansal, Yuval, et autres
Publié: (2025)

Beyond QA Pairs: Assessing Parameter-Efficient Fine-Tuning for Fact Embedding in LLMs
par: Ratnakar, Shivam, et autres
Publié: (2025)

Evidence-backed Fact Checking using RAG and Few-Shot In-Context Learning with LLMs
par: Singhal, Ronit, et autres
Publié: (2024)

Molecular Facts: Desiderata for Decontextualization in LLM Fact Verification
par: Gunjal, Anisha, et autres
Publié: (2024)

Not All Needles Are Found: How Fact Distribution and Don't Make It Up Prompts Shape Literal Extraction, Logical Inference, and Hallucination Risks in Long-Context LLMs
par: Ebrahimzadeh, Amirali, et autres
Publié: (2026)

FactCG: Enhancing Fact Checkers with Graph-Based Multi-Hop Data
par: Lei, Deren, et autres
Publié: (2025)

Latent Adversarial Training Improves Robustness to Persistent Harmful Behaviors in LLMs
par: Sheshadri, Abhay, et autres
Publié: (2024)

Knowing the Facts but Choosing the Shortcut: Understanding How Large Language Models Compare Entities
par: Lehmann, Hans Hergen, et autres
Publié: (2025)

KG-CRAFT: Knowledge Graph-based Contrastive Reasoning with LLMs for Enhancing Automated Fact-checking
par: Lourenço, Vítor N., et autres
Publié: (2026)