arXiv (All)
open access
Taming CoT Obfuscation in VLMs: From Mechanistic Evidence to Activation Enforcement
Reinforcement learning (RL) improves reasoning in vision-language models (VLMs) but can induce chain-of-thought (CoT) obfuscation: an operational, non-intentional outcome where task reward or accuracy rises while traces become less grounded and monitorable. Prior work largely documents this decay be
Conceptio Open Archive
Analyze & explore related research in Conceptio
Connect this record directly into your research workflow or AI agent. Explore citations across open repositories, query full text, and export structured proof bundles.
This document is indexed with metadata only: full text is not available in the archive for this record. Open the official source ↗
Related documents
- Large Language Models Hack Rewards, and Society#1129288
- Length Penalties Make Chain-of-Thought Less Monitorable#1050911
- Official Review · ICLR 2026#787021
- When a Kindergartener Solves Calculus: Measuring Capability Leakage in Role-Prompted Reasoning Models#1145013
- Beyond Changing the Answer: Warranted Downstream Correction in Large Language Models A Behavioral Framework with Evidence from Open-Ended Causal Investigations#694017
Record · ID 1050426
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Retrieved via Conceptio: every document is proof-bundled with source, license, and retrieval metadata.
Bring this data into your agent: Connect via MCP or API · 20 free credits to start · no card required
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive
Read the Docs → · View CLI source on GitHub · Conceptio Open Knowledge Archive