Saved in:
Bibliographic Details
Main Authors: Huntsman, Steve, Thomas, Jewell
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2502.13953
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915452320481280
author Huntsman, Steve
Thomas, Jewell
author_facet Huntsman, Steve
Thomas, Jewell
contents We devise an algorithm to generate propositions that objectively instantiate graphs supporting coherence-driven inference. We also benchmark the ability of large language models (LLMs) to reconstruct coherence graphs from (a simple transformation of) propositions expressed in natural language, with promising results from a single prompt to reasoning-optimized LLMs. For example, o1/3/4-mini achieve perfect reconstruction half of the time on sparse graphs. Coherence-driven inference on consistency evaluations by LLMs may advance machine cognition capabilities.
format Preprint
id arxiv_https___arxiv_org_abs_2502_13953
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Benchmarking graph construction by large language models for coherence-driven inference
Huntsman, Steve
Thomas, Jewell
Artificial Intelligence
We devise an algorithm to generate propositions that objectively instantiate graphs supporting coherence-driven inference. We also benchmark the ability of large language models (LLMs) to reconstruct coherence graphs from (a simple transformation of) propositions expressed in natural language, with promising results from a single prompt to reasoning-optimized LLMs. For example, o1/3/4-mini achieve perfect reconstruction half of the time on sparse graphs. Coherence-driven inference on consistency evaluations by LLMs may advance machine cognition capabilities.
title Benchmarking graph construction by large language models for coherence-driven inference
topic Artificial Intelligence
url https://arxiv.org/abs/2502.13953