# Large Language Model Critics for Execution-Free Evaluation of Code Changes

arXiv: https://arxiv.org/abs/2501.16655 · HTML: https://mlsanity.com/paper/2501.16655/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- First posted on arXiv 28 Jan 2025.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)

## Built on by (most cited first, 7 of 7)

- [GUI Agents for Continual Game Generation](https://mlsanity.com/paper/2605.28258/) (2026)
- [Evolutionary Perspectives on the Evaluation of LLM-Based AI Agents: A Comprehensive Survey](https://mlsanity.com/paper/2506.11102/) (2025)
- [Agentic Software Issue Resolution with Large Language Models: A Survey](https://mlsanity.com/paper/2512.22256/) (2025)
- [Steer, Don't Solve: Training Small Critic Models for Large Code Agents](https://mlsanity.com/paper/2606.21811/) (2026)
- [Themis: Training Robust Multilingual Code Reward Models for Flexible Multi-Criteria Scoring](https://mlsanity.com/paper/2605.00754/) (2026)
- [Groundability, Not Scale Alone: When Weak Reviewers Can Audit Strong Coding Agents](https://mlsanity.com/paper/2610.01023/) (2026)
- [Can LLMs Fix It Without Code? Toward Automated Verification of No-Code Bug Fixes](https://mlsanity.com/paper/2610.11963/) (2026)

## Abstract (arXiv)

Large language models (LLMs) offer a promising way forward for automating software engineering tasks, such as bug fixes, feature additions, etc., via multi-step LLM-based agentic workflows. However, existing metrics for evaluating such workflows, mainly build status and occasionally log analysis, are too sparse and limited in providing the information needed to assess the quality of changes made. In this work, we designed LLM-based critics to derive well-structured and rigorous intermediate/step-level, execution-free evaluation proxies for repo-level code changes. Importantly, we assume access to the gold test patch for the problem (i.e., reference-aware) to assess both semantics and executability of generated patches. With the gold test patch as a reference, we predict executability of all editing locations with an F1 score of 91.6%, aggregating which, we can predict the build status in 84.8% of the instances in SWE-bench. In particular, such an execution-focused LLM critic outperforms other reference-free and reference-aware LLM critics by 38.9% to 72.5%. Moreover, we demonstrate the usefulness of such a reference-aware framework in comparing patches generated by different agentic workflows. Finally, we open-source the library developed for this project, which allows further usage for either other agentic workflows or other benchmarks. The source code is available at https://github.com/amazon-science/code-agent-eval.