Aletheia: Quantifying Cognitive Conviction in Reasoning Models via Regularized Inverse Confusion Matrix

📝 Original Info

• Title: Aletheia: Quantifying Cognitive Conviction in Reasoning Models via Regularized Inverse Confusion Matrix
• ArXiv ID: 2601.01532
• Date: 2026-01-04
• Authors: Fanzhe Fu

📝 Abstract

In the progressive journey toward Artificial General Intelligence (AGI), current evaluation paradigms face an epistemological crisis. Static benchmarks measure knowledge breadth but fail to quantify the depth of belief. While Simhi et al. [2025] defined the CHOKE phenomenon in standard QA, we extend this framework to quantify Cognitive Conviction in System 2 reasoning models. We propose Project Aletheia, a cognitive physics framework that employs Tikhonov Regularization to invert the judge's confusion matrix. To validate this methodology without relying on opaque private data, we implement a Synthetic Proxy Protocol. Our preliminary pilot study on 2025 baselines (e.g., DeepSeek-R1, OpenAI o1) suggests that while reasoning models act as a "cognitive buffer," they may exhibit Defensive Over-Thinking under adversarial pressure. Furthermore, we introduce the Aligned Conviction Score (S aligned ) to verify that conviction does not compromise safety. This work serves as a blueprint for measuring AI scientific integrity.

📄 Full Content

...(본문 내용이 길어 생략되었습니다. 사이트에서 전문을 확인해 주세요.)