When Confidence Rises Too Early: Detecting Shortcut Reasoning via Premature Answer Commitment
cs.CL
Submitted: 2026-09-28
Updated: 2026-09-28
Terminology
Sources
- DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
- Evaluating Large Language Models Trained on Code
- OpenAI o1 System Card
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- Qwen3 Technical Report
- Reasoning Models Don't Always Say What They Think
- Is It Thinking or Cheating? Detecting Implicit Reward Hacking by Measuring Reasoning Effort
- Making Slow Thinking Faster: Compressing LLM Chain-of-Thought via Step Entropy
- Natural Emergent Misalignment from Reward Hacking in Production RL
- Truthful or Fabricated? Using Causal Attribution to Mitigate Reward Hacking in Explanations
- School of Reward Hacks: Hacking harmless tasks generalizes to misaligned behavior in LLMs
- Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
- CoT Red-Handed: Stress Testing Chain-of-Thought Monitoring
- When Chain of Thought is Necessary, Language Models Struggle to Evade Monitors
- Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
- RewardBench 2: Advancing Reward Model Evaluation
- Probing the Trajectories of Reasoning Traces in Large Language Models
- Language Models (Mostly) Know What They Know
- Skywork-Reward-V2: Scaling Preference Data Curation via Human-AI Synergy
- Measuring Faithfulness in Chain-of-Thought Reasoning
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