SGCR: A Specification-Grounded Framework for Trustworthy LLM Code Review

TL;DR

SGCR is a framework that enhances trustworthiness in LLM-based code review by grounding models in human specifications, combining deterministic rule compliance with heuristic issue detection, leading to significant practical improvements.

Contribution

This paper introduces a novel specification-grounded architecture for LLM code review, integrating explicit rule adherence with implicit issue discovery to improve reliability.

Findings

42% developer adoption rate in industry

90.9% relative improvement over baseline LLM

Effective bridging of generative power and reliability demands

Abstract

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.