Skip to content
AI News HubLIVE
Original source2 min read

Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty

Summary

The paper proposes a "structural uncertainty" framework that evaluates logical reasoning consistency by analyzing the stability of a model's self-preference rankings over multiple generated reasoning candidates. Experiments across five LLMs and eight benchmarks show that structural signals complement answer dispersion: combining both improves identification of unreliable instances on logical and mathematical reasoning tasks, while on factual retrieval the structural signal collapses toward uniformity, revealing a regime boundary where reasoning-level consistency evaluation is uninformative.

SourcearXiv AIAuthor: Baishali Chaudhury, Mengdie Flora Wang, Hyunji Hayley Park, Rahul Ghosh, Sungmin Hong, Jae Oh Woo
Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty
Report an error

The correction channel is not available yet. You can copy the article reference below for later.

Correction instructions
Read article

[2606.17312] Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty

[Submitted on 15 Jun 2026]

Title:Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty

View a PDF of the paper titled Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty, by Baishali Chaudhury and 5 other authors

View PDF HTML (experimental)

Abstract:Large language models can arrive at the same answer through reasoning paths that are unstable, contradictory, or difficult to rank consistently -- a failure mode especially prevalent in multi-step deductive reasoning. Existing methods assess reliability primarily through output dispersion -- measuring how much sampled answers differ -- but this discards a complementary signal: whether the model can consistently rank competing reasoning candidates. We propose structural uncertainty, a consistency-aware framework derived from the stability of self-preference-induced rankings over sampled reasoning solutions. Given a query, we generate multiple candidate solutions and ask the model to judge pairwise preferences among its own outputs. We aggregate self-preferences into ranking distributions via Bradley-Terry modeling with PageRank, and decompose the signal into two entropy-based components: across-trial ranking instability and within-trial candidate ambiguity. Across five LLMs and eight benchmarks, structural signals provide information complementary to answer dispersion: on logical and mathematical reasoning tasks, the combination improves identification of unreliable instances, while on factual retrieval the structural signal collapses toward uniformity, diagnosing a regime boundary where reasoning-level consistency evaluation is uninformative. The two components relate differently to accuracy: within-trial ambiguity correlates positively with correctness -- consistent with settings where multiple plausible solution paths remain competitive -- while across-trial instability correlates negatively, signaling unreliable reasoning. Structural uncertainty is best understood not as a universal confidence estimator, but as a regime-sensitive evaluator of logical reasoning consistency.

Comments: Published at ICLR 2026 Workshop on Logical Reasoning of Large Language Models. Accepted as best paper

Subjects:

Artificial Intelligence (cs.AI)

Cite as: arXiv:2606.17312 [cs.AI]

(or arXiv:2606.17312v1 [cs.AI] for this version)

https://doi.org/10.48550/arXiv.2606.17312

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Baishali Chaudhury [view email] [v1] Mon, 15 Jun 2026 21:39:47 UTC (8,497 KB)

Full-text links:

Access Paper:

View a PDF of the paper titled Quantifying Consistency in LLM Logical Reasoning via Structural Uncertainty, by Baishali Chaudhury and 5 other authors

View PDF

HTML (experimental)

TeX Source

view license

Current browse context:

cs.AI

new | recent | 2026-06

Change to browse by:

cs

References & Citations

NASA ADS

Google Scholar

Semantic Scholar

Loading...

Data provided by:

Bibliographic Tools

Bibliographic and Citation Tools

Bibliographic Explorer Toggle

Bibliographic Explorer (What is the Explorer?)

Connected Papers Toggle

Connected Papers (What is Connected Papers?)

Litmaps Toggle

Litmaps (What is Litmaps?)

scite.ai Toggle

scite Smart Citations (What are Smart Citations?)

Code, Data, Media

Code, Data and Media Associated with this Article

alphaXiv Toggle

alphaXiv (What is alphaXiv?)

Links to Code Toggle

CatalyzeX Code Finder for Papers (What is CatalyzeX?)

DagsHub Toggle

DagsHub (What is DagsHub?)

GotitPub Toggle

Gotit.pub (What is GotitPub?)

Huggingface Toggle

Hugging Face (What is Huggingface?)

ScienceCast Toggle

ScienceCast (What is ScienceCast?)

Demos

Demos

Replicate Toggle

Replicate (What is Replicate?)

Spaces Toggle

Hugging Face Spaces (What is Spaces?)

Spaces Toggle

TXYZ.AI (What is TXYZ.AI?)

Related Papers

Recommenders and Search Tools

Link to Influence Flower

Influence Flower (What are Influence Flowers?)

Core recommender toggle

CORE Recommender (What is CORE?)

Author

Venue

Institution

Topic

About arXivLabs

arXivLabs: experimental projects with community collaborators

arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.

Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.

Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.

Which authors of this paper are endorsers? | Disable MathJax (What is MathJax?)

Key points and analysis

Article intelligence

InvestorsAdvanced

Key points

  • Structural uncertainty quantifies reasoning consistency by having the model rank its own outputs via pairwise preferences, complementing traditional answer dispersion metrics.
  • Ranking distributions are aggregated using Bradley-Terry modeling with PageRank and decomposed into across-trial ranking instability and within-trial candidate ambiguity.
  • Within-trial ambiguity correlates positively with correctness, while across-trial instability correlates negatively, corresponding to multi-solution competition and unreliable reasoning respectively.
  • On factual retrieval tasks, the structural signal collapses toward uniformity, indicating the method is primarily suited for logical and mathematical reasoning scenarios.

Highlights and analysis are generated automatically and may contain errors. Check the original source.