Efficient Reasoning Training Does Not Always Harm CoT Faithfulness and Monitorability
cs.AI
Submitted: 2026-10-02
Updated: 2026-10-02
Sources
- Monitoring Reasoning Models for Misbehavior and the Risks of Promoting Obfuscation
- Reasoning Models Don't Always Say What They Think
- Chain-of-Thought Faithfulness of Reasoning Models Varies with Where and How Preference Cues Are Delivered
- Gemma 3 Technical Report
- Monitoring Monitorability
- What Disease does this Patient Have? A Large-scale Open Domain Question Answering Dataset from Medical Exams
- Kimi k1.5: Scaling Reinforcement Learning with LLMs
- Chain of Thought Monitorability: A New and Fragile Opportunity for AI Safety
- Reasoning Under Pressure: How do Training Incentives Influence Chain-of-Thought Monitorability?
- Olmo 3
- gpt-oss-120b & gpt-oss-20b Model Card
- Addressing cognitive bias in medical language models
- Qwen3 Technical Report
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems