When Chain-of-Thought Fails, the Solution Hides in the Hidden States

arXiv:2604.23351 · cs.CL, cs.AI, cs.LG · Submitted 2026-04-25 · Read on arXiv

cs.CL, cs.AI, cs.LG

Submitted: 2026-04-25

Updated: 2026-09-02

Code: https://github.com/hmehrafarin/when-cot-fails

Terminology

• Trace — 28× in this paper · explained in 2 episodes
• Accuracy — 27× in this paper · explained in 1 episode
• Activation Patching — 7× in this paper · explained in 2 episodes
• Interpretability — 7× in this paper · explained in 15 episodes
• Transformer — 7× in this paper · explained in 2 episodes
• Benchmark — 4× in this paper · explained in 2 episodes
• Chain-of-Thought (CoT) — 3× in this paper · explained in 4 episodes
• Chain-of-Thought Prompting — 3× in this paper · explained in 2 episodes
• Consistency — 3× in this paper · explained in 3 episodes
• Self-consistency — 3× in this paper · explained in 2 episodes
• Reasoning Traces — 2× in this paper · explained in 3 episodes
• Abstraction — 1× in this paper · explained in 1 episode
• Adam — 1× in this paper · explained in 2 episodes
• Chain-of-Thought (CoT) Prompting — 1× in this paper · explained in 3 episodes
• Latent Reasoning — 1× in this paper · explained in 2 episodes
• Pairwise Comparisons — 1× in this paper · explained in 2 episodes
• Structured Reasoning — 1× in this paper · explained in 3 episodes
• Verification — 1× in this paper · explained in 2 episodes
• Verifier — 1× in this paper · explained in 2 episodes

Sources

• How does Chain of Thought Think? Mechanistic Interpretability of Chain-of-Thought Reasoning with Sparse Autoencoding
• The Llama 3 Herd of Models
• Qwen2.5 Technical Report
• Measuring Faithfulness in Chain-of-Thought Reasoning
• Can Aha Moments Be Fake? Towards Quantifying Decorative and True Thinking in Chain-of-Thought

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