LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification
#LogicGraph #Multi-path reasoning #Large language models #Neuro-symbolic framework #Logical reasoning #AI evaluation #Convergent reasoning
📌 Key Takeaways
- LogicGraph is the first benchmark specifically designed to evaluate multi-path logical reasoning in AI systems
- Current evaluations focus on convergent reasoning with single correct proofs, unlike real-world problems
- The benchmark uses a neuro-symbolic framework with backward logic generation and semantic instantiation
- Experiments reveal AI models commit early to single reasoning paths, with performance gaps increasing with complexity
📖 Full Retelling
Researchers led by Yanrui Wu and 7 other authors introduced LogicGraph, a new benchmark for evaluating multi-path logical reasoning in AI, published on arXiv on February 24, 2026, addressing the limitation in current large language model evaluations that only test convergent reasoning requiring a single correct proof. The benchmark represents a significant advancement in AI evaluation methodology, as most existing assessments focus on problems with only one valid solution path, while many real-world reasoning problems allow for multiple valid derivations. LogicGraph was constructed through a neuro-symbolic framework that utilizes backward logic generation and semantic instantiation to create solver-verified reasoning problems featuring high-depth multi-path reasoning and inherent logical distractions. Each benchmark instance is associated with an exhaustive set of minimal proofs, enabling comprehensive evaluation of AI reasoning capabilities. The researchers also developed a reference-free evaluation framework to rigorously assess model performance in both convergent and divergent reasoning regimes. Experiments conducted on state-of-the-art language models revealed a common limitation: models tend to commit early to a single reasoning path and fail to explore alternative approaches, with the coverage gap between human and AI reasoning growing substantially as reasoning depth increases. This divergence gap exposed by LogicGraph provides actionable insights to motivate future improvements in AI reasoning systems.
🏷️ Themes
Artificial Intelligence, Logical Reasoning, Benchmark Development, Neuro-Symbolic Systems
📚 Related People & Topics
Logical reasoning
Process of drawing correct inferences
Logical reasoning is a mental activity that aims to arrive at a conclusion in a rigorous way. It happens in the form of inferences or arguments by starting from a set of premises and reasoning to a conclusion supported by these premises. The premises and the conclusion are propositions, i.e.
Large language model
Type of machine learning model
A large language model (LLM) is a language model trained with self-supervised machine learning on a vast amount of text, designed for natural language processing tasks, especially language generation. The largest and most capable LLMs are generative pre-trained transformers (GPTs) that provide the c...
Entity Intersection Graph
Connections for Logical reasoning:
View full profileOriginal Source
--> Computer Science > Artificial Intelligence arXiv:2602.21044 [Submitted on 24 Feb 2026] Title: LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification Authors: Yanrui Wu , Lingling Zhang , Xinyu Zhang , Jiayu Chang , Pengyu Li , Xu Jiang , Jingtao Hu , Jun Liu View a PDF of the paper titled LogicGraph : Benchmarking Multi-Path Logical Reasoning via Neuro-Symbolic Generation and Verification, by Yanrui Wu and 7 other authors View PDF HTML Abstract: Evaluations of large language models 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 this https URL . Comments: 24 pages, 17 figures Subjects: Artificial Intelligence (cs.AI) Cite as: arXiv:2602.21044 [cs.AI] (or arXiv:2602.21044v1 [cs.AI] for this version) https://doi.org/10.48550/arXiv.2602.21044 Focus to ...
Read full article at source