AI Paper Atlas

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

The Geometry of Reasoning: Flowing Logics in Representation Space

Yufa Zhou, Yixiao Wang, Xunjian Yin, Shuyan Zhou, Anru Zhang

ICLR 2026 · Large language models › Agents, code and math · approach: Inductive and abstract reasoning evaluation

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
Compare first- and second-order geometry (velocity, curvature) instead of raw positions when grouping flows by logicReasoning-flow similarity · Logic dataset, Qwen3 0.6BSimilarity: position 0.26 vs velocity 0.17 / curvature 0.53 for logic; topic 0.07 / 0.11Improved
Randomly shuffle the order of logical steps (ablation)Qwen3 0.6B reasoning flows · Logic datasetVelocity and curvature similarity: velocity 0.17 -> 0.02; curvature 0.53 -> 0.02Hurt
Randomly shuffle the order of logical steps (ablation)Qwen3 0.6B reasoning flows · Logic datasetPosition similarity: language 0.85 -> 0.81No clear effect
Scale model size (Qwen3 0.6B to 1.7B to 4B) (ablation)Reasoning-flow similarity · Logic datasetVelocity and curvature similarity: velocity 0.17, 0.19, 0.16No clear effect
Switch model family (Qwen1.5, Qwen2, Qwen3, LLaMA3) (ablation)Reasoning-flow similarity · Logic datasetCurvature similarity: 0.52, 0.52, 0.53, 0.58No clear effect

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