Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Kai, Mao, Bingcheng, Jia, Shuai, Ding, Yujie, Han, Dongming, Ma, Tianyi, Cao, Bin
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:https://arxiv.org/abs/2512.17540
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866910001582309376
author Wang, Kai
Mao, Bingcheng
Jia, Shuai
Ding, Yujie
Han, Dongming
Ma, Tianyi
Cao, Bin
author_facet Wang, Kai
Mao, Bingcheng
Jia, Shuai
Ding, Yujie
Han, Dongming
Ma, Tianyi
Cao, Bin
contents Automating code review with Large Language Models (LLMs) shows immense promise, yet practical adoption is hampered by their lack of reliability, context-awareness, and control. To address this, we propose Specification-Grounded Code Review (SGCR), a framework that grounds LLMs in human-authored specifications to produce trustworthy and relevant feedback. SGCR features a novel dual-pathway architecture: an explicit path ensures deterministic compliance with predefined rules derived from these specifications, while an implicit path heuristically discovers and verifies issues beyond those rules. Deployed in a live industrial environment at HiThink Research, SGCR's suggestions achieved a 42% developer adoption rate-a 90.9% relative improvement over a baseline LLM (22%). Our work demonstrates that specification-grounding is a powerful paradigm for bridging the gap between the generative power of LLMs and the rigorous reliability demands of software engineering.
format Preprint
id arxiv_https___arxiv_org_abs_2512_17540
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle SGCR: A Specification-Grounded Framework for Trustworthy LLM Code Review
Wang, Kai
Mao, Bingcheng
Jia, Shuai
Ding, Yujie
Han, Dongming
Ma, Tianyi
Cao, Bin
Software Engineering
Automating code review with Large Language Models (LLMs) shows immense promise, yet practical adoption is hampered by their lack of reliability, context-awareness, and control. To address this, we propose Specification-Grounded Code Review (SGCR), a framework that grounds LLMs in human-authored specifications to produce trustworthy and relevant feedback. SGCR features a novel dual-pathway architecture: an explicit path ensures deterministic compliance with predefined rules derived from these specifications, while an implicit path heuristically discovers and verifies issues beyond those rules. Deployed in a live industrial environment at HiThink Research, SGCR's suggestions achieved a 42% developer adoption rate-a 90.9% relative improvement over a baseline LLM (22%). Our work demonstrates that specification-grounding is a powerful paradigm for bridging the gap between the generative power of LLMs and the rigorous reliability demands of software engineering.
title SGCR: A Specification-Grounded Framework for Trustworthy LLM Code Review
topic Software Engineering
url https://arxiv.org/abs/2512.17540