AI Paper Atlas

Research map / Large language models / Agents, code and math

Decoupling Knowledge and Reasoning in LLMs: An Exploration Using Cognitive Dual-System Theory

Mutian Yang, Jiandong Gao, Ji Wu

AAAI 2026 · Large language models › Agents, code and math · approach: Mathematical and structural reasoning benchmarks · 5 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
slow thinking (CoT) instead of fast thinkingQwen 1.5B · MMLUaccuracy: 53.9 -> 50.0Hurt
slow thinking (CoT) instead of fast thinkingQwen 7B · MMLUaccuracy: 67.8 -> 71.8Improved
slow thinking (CoT) instead of fast thinkingLLaMA 1B · MMLUaccuracy: 35.1 -> 35.1No clear effect
slow thinking in mathematics domain15 LLMs · MMLU Mathematicsaccuracy: average delta 19.0Improved
slow thinking in political science domain15 LLMs · MMLU Political Scienceaccuracy: average delta -3.0Hurt
scale LLaMA from 1B to 70Boverthinking rate · MMLUro (%): 52.6 -> 7.2Improved
scale LLaMA from 1B to 70Bcorrection rate · MMLUrc (%): 29.4 -> 38.1Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy