CoT is Not the Chain of Truth: An Empirical Internal Analysis of Reasoning LLMs for Fake News Generation
cs.CL
Submitted: 2026-02-04
Updated: 2026-09-08
Comments: Accepted at the 43rd International Conference on Machine Learning (ICML 2026)
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Attention Mechanisms Perspective: Exploring LLM Processing of Graph-Structured Data
- Training Compute-Optimal Large Language Models
- CoT Red-Handed: Stress Testing Chain-of-Thought Monitoring
- Qwen Technical Report
- Developing Story: Case Studies of Generative AI's Use in Journalism
- HiddenDetect: Detecting Jailbreak Attacks against Large Vision-Language Models via Monitoring Hidden States
- How Smooth Is Attention?
- Universal Model Routing for Efficient LLM Inference
- Reasoning Models Don't Always Say What They Think
- What Does BERT Look At? An Analysis of BERT's Attention
- Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
- Multi-step Jailbreaking Privacy Attacks on ChatGPT
- Robust Fake News Detection using Large Language Models under Adversarial Sentiment Attacks
- Measuring Chain-of-Thought Monitorability Through Faithfulness and Verbosity
- X-Teaming: Multi-Turn Jailbreaks and Defenses with Adaptive Multi-Agents
- Analyzing Multi-Head Self-Attention: Specialized Heads Do the Heavy Lifting, the Rest Can Be Pruned
- GLU Variants Improve Transformer
- Don't Take Things Out of Context: Attention Intervention for Enhancing Chain-of-Thought Reasoning in Large Language Models
- LLMs Encode Harmfulness and Refusal Separately
Related papers
- Exploring Solution Divergence and Its Effect on Large Language Model Problem Solving
- Ishigaki-IDS-Bench: A Benchmark for Generating Information Delivery Specification from BIM Information Requirements
- Subliminal Steering: Stronger Encoding of Hidden Signals
- MedStruct-S: A Benchmark for Key Discovery, Key-Conditioned QA and Semi-Structured Extraction from OCR Clinical Reports
- The End of Transformers? On Challenging Attention and the Rise of Sub-Quadratic Architectures
- Untangling the Mechanisms of Misleading Context in Medical Question Answering