AI Paper Atlas

Research map / Large language models / Reasoning and chain of thought

The Geometry of Reasoning: Self-Evaluation via Layerwise Trajectory Evolution

Jinhe Bi, Danqi Yan, Yifan Wang, Wenke Huang, Haokun Chen, Guancheng Wan, Mang Ye, Xun Xiao, Hinrich Schuetze, Volker Tresp, Yunpu Ma

ICML 2026 · Large language models › Reasoning and chain of thought · approach: Long CoT dynamics and efficiency

Read the paperCheck your idea against 137,217 papers

AI-extracted summary. Written by AI from the paper's full text. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Proposes GeoR, a method for letting large reasoning models self-evaluate the correctness of their chain-of-thought trajectories by analyzing the layerwise evolution of hidden representations, without external supervision.

Problem

Large reasoning models need to judge whether their own CoT reasoning is correct without ground-truth labels, but existing methods rely on labels or shallow output probabilities and ignore how the reasoning trajectory evolves across layers.

Approach

Treats the layerwise evolution of the reasoning trajectory inside the model as a geometric signal for self-evaluation, rather than relying on ground-truth labels or final-output token probabilities.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
model evolution along layers instead of along tokens (ablation)GeoR · GSM8K (R1-Qwen-1.5B)AUROC: 51.84 -> 77.28Improved
model evolution along layers instead of along tokens (ablation)GeoR · GSM8K (R1-LLaMA-8B)AUROC: 48.93 -> 68.12Improved
average hidden states over reasoning tokens instead of using last CoT token (ablation)GeoR · GSM8K (R1-Qwen-1.5B)AUROC: 71.54 -> 77.28Improved
average over reasoning tokens instead of the whole output (ablation)GeoR · GSM8K (R1-Qwen-14B)AUROC: 55.64 -> 61.64Improved
normalize geometric path by total displacement (ablation)GeoR · GSM8K (R1-Qwen-7B)AUROC: 75.10 -> 76.32Improved
add difficulty-aware calibration H to geometric evolution (ablation)GeoR · GSM8K (R1-Qwen-7B)AUROCImproved
use GeoR instead of best baseline CoE-RCoE-R · GSM8K (R1-Qwen-1.5B)AUROC: 74.17 -> 77.28Improved
use GeoR instead of EntropyEntropy · TheoremQA (R1-Qwen-14B)AUROC: 52.62 -> 52.65No clear effect
use GeoR instead of MaxProbMaxProb · CommonsenseQA (R1-LLaMA-8B)AUROC: 74.71 -> 74.71No clear effect
use GeoR instead of CoE-R in repeated promptsCoE-R · GSM8K (3 prompts, R1-Qwen-7B)AUROC: 71.3 -> 76.8Improved
use GeoR instead of sampling-based Internal ConsistencyInternal Consistency · GSM8K (R1-LLaMA-8B)latency: 151.14 s -> 1.2 × 10−3 sImproved

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