Saved in:
Bibliographic Details
Main Authors: Chen, Kang, Nian, Junjie, Cao, Yixin, Jiang, Yugang
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2605.14619
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917495510663168
author Chen, Kang
Nian, Junjie
Cao, Yixin
Jiang, Yugang
author_facet Chen, Kang
Nian, Junjie
Cao, Yixin
Jiang, Yugang
contents Multi-run chain-of-thought reasoning is usually collapsed to final-answer aggregates, which discard howsampled trajectories share, split, and rejoin through intermediate computation. We propose SliceGraph, a post-hoc problem-model-cell graph built by mutual-kNN over sparse activation-key Jaccard similarity between CoT slices, and treat it as a measurement object for process geometry rather than as a decoding program. Across sampled CoT ensembles from three primary 4B/8B models on math and science benchmarks, blinded annotation supports SliceGraph biconnected components as shared reasoning-state units and process families as within-family strategy-coherent route units. In 85.5% of 954 problem-model cells, correct CoTs sharing the same normalized answer split into multiple process families; among cells with at least two such runs, 76.6% of run pairs are cross-family on average. We call such same-answer, family-divergent correct trajectories process isomers. A label-seeded reward field provides a separate value-landscape layer: success-associated regions often split into disconnected high-value cores, and route families specialize over these core footprints rather than merely duplicating one another. A typed-state transition analysis further shows that process families navigate the same atlas with distinct transition kernels under matched null controls. Representation ablations, a cross-architecture replication, and two cross-scale replications support the robustness of the route-family scaffold, showing that final-answer aggregation overlooks this structured multi-route process geometry.
format Preprint
id arxiv_https___arxiv_org_abs_2605_14619
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SliceGraph: Mapping Process Isomers in Multi-Run Chain-of-Thought Reasoning
Chen, Kang
Nian, Junjie
Cao, Yixin
Jiang, Yugang
Artificial Intelligence
Multi-run chain-of-thought reasoning is usually collapsed to final-answer aggregates, which discard howsampled trajectories share, split, and rejoin through intermediate computation. We propose SliceGraph, a post-hoc problem-model-cell graph built by mutual-kNN over sparse activation-key Jaccard similarity between CoT slices, and treat it as a measurement object for process geometry rather than as a decoding program. Across sampled CoT ensembles from three primary 4B/8B models on math and science benchmarks, blinded annotation supports SliceGraph biconnected components as shared reasoning-state units and process families as within-family strategy-coherent route units. In 85.5% of 954 problem-model cells, correct CoTs sharing the same normalized answer split into multiple process families; among cells with at least two such runs, 76.6% of run pairs are cross-family on average. We call such same-answer, family-divergent correct trajectories process isomers. A label-seeded reward field provides a separate value-landscape layer: success-associated regions often split into disconnected high-value cores, and route families specialize over these core footprints rather than merely duplicating one another. A typed-state transition analysis further shows that process families navigate the same atlas with distinct transition kernels under matched null controls. Representation ablations, a cross-architecture replication, and two cross-scale replications support the robustness of the route-family scaffold, showing that final-answer aggregation overlooks this structured multi-route process geometry.
title SliceGraph: Mapping Process Isomers in Multi-Run Chain-of-Thought Reasoning
topic Artificial Intelligence
url https://arxiv.org/abs/2605.14619