LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification

TL;DR

LogicGraph introduces a benchmark for multi-path logical reasoning in language models, highlighting their tendency to focus on single solutions and revealing the need for models to explore diverse reasoning paths in complex problems.

Contribution

This work presents the first benchmark for multi-path logical reasoning, using a neuro-symbolic framework to generate and verify diverse reasoning paths, and proposes a new evaluation method for model performance.

Findings

Models tend to commit early to a single reasoning route.

Coverage gap increases with reasoning depth.

LogicGraph exposes divergence in model reasoning capabilities.

Abstract

Evaluations of large language models (LLMs) primarily emphasize convergent logical reasoning, where success is defined by producing a single correct proof. However, many real-world reasoning problems admit multiple valid derivations, requiring models to explore diverse logical paths rather than committing to one route. To address this limitation, we introduce LogicGraph, the first benchmark aimed to systematically evaluate multi-path logical reasoning, constructed via a neuro-symbolic framework that leverages backward logic generation and semantic instantiation. This pipeline yields solver-verified reasoning problems formalized by high-depth multi-path reasoning and inherent logical distractions, where each instance is associated with an exhaustive set of minimal proofs. We further propose a reference-free evaluation framework to rigorously assess model performance in both convergent and divergent regimes. Experiments on state-of-the-art language models reveal a common limitation: models tend to commit early to a single route and fail to explore alternatives, and the coverage gap grows substantially with reasoning depth. LogicGraph exposes this divergence gap and provides actionable insights to motivate future improvements. Our code and data will be released at https://github.com/kkkkarry/LogicGraph.