Salvato in:
Dettagli Bibliografici
Autori principali: Kalyanpur, Aditya, Saravanakumar, Kailash Karthik, Barres, Victor, Chu-Carroll, Jennifer, Melville, David, Ferrucci, David
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:https://arxiv.org/abs/2406.17663
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916329999564800
author Kalyanpur, Aditya
Saravanakumar, Kailash Karthik
Barres, Victor
Chu-Carroll, Jennifer
Melville, David
Ferrucci, David
author_facet Kalyanpur, Aditya
Saravanakumar, Kailash Karthik
Barres, Victor
Chu-Carroll, Jennifer
Melville, David
Ferrucci, David
contents We introduce LLM-ARC, a neuro-symbolic framework designed to enhance the logical reasoning capabilities of Large Language Models (LLMs), by combining them with an Automated Reasoning Critic (ARC). LLM-ARC employs an Actor-Critic method where the LLM Actor generates declarative logic programs along with tests for semantic correctness, while the Automated Reasoning Critic evaluates the code, runs the tests and provides feedback on test failures for iterative refinement. Implemented using Answer Set Programming (ASP), LLM-ARC achieves a new state-of-the-art accuracy of 88.32% on the FOLIO benchmark which tests complex logical reasoning capabilities. Our experiments demonstrate significant improvements over LLM-only baselines, highlighting the importance of logic test generation and iterative self-refinement. We achieve our best result using a fully automated self-supervised training loop where the Actor is trained on end-to-end dialog traces with Critic feedback. We discuss potential enhancements and provide a detailed error analysis, showcasing the robustness and efficacy of LLM-ARC for complex natural language reasoning tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2406_17663
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle LLM-ARC: Enhancing LLMs with an Automated Reasoning Critic
Kalyanpur, Aditya
Saravanakumar, Kailash Karthik
Barres, Victor
Chu-Carroll, Jennifer
Melville, David
Ferrucci, David
Computation and Language
Artificial Intelligence
Logic in Computer Science
We introduce LLM-ARC, a neuro-symbolic framework designed to enhance the logical reasoning capabilities of Large Language Models (LLMs), by combining them with an Automated Reasoning Critic (ARC). LLM-ARC employs an Actor-Critic method where the LLM Actor generates declarative logic programs along with tests for semantic correctness, while the Automated Reasoning Critic evaluates the code, runs the tests and provides feedback on test failures for iterative refinement. Implemented using Answer Set Programming (ASP), LLM-ARC achieves a new state-of-the-art accuracy of 88.32% on the FOLIO benchmark which tests complex logical reasoning capabilities. Our experiments demonstrate significant improvements over LLM-only baselines, highlighting the importance of logic test generation and iterative self-refinement. We achieve our best result using a fully automated self-supervised training loop where the Actor is trained on end-to-end dialog traces with Critic feedback. We discuss potential enhancements and provide a detailed error analysis, showcasing the robustness and efficacy of LLM-ARC for complex natural language reasoning tasks.
title LLM-ARC: Enhancing LLMs with an Automated Reasoning Critic
topic Computation and Language
Artificial Intelligence
Logic in Computer Science
url https://arxiv.org/abs/2406.17663