# SynthFix: Adaptive Neuro-Symbolic Code Vulnerability Repair

arXiv: https://arxiv.org/abs/2604.17184 · HTML: https://mlsanity.com/paper/2604.17184/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 27 of its 72 references that are in the map.
- First posted on arXiv 19 Apr 2026.

## Builds on (most cited first, 27 of 27)

- [Proximal Policy Optimization Algorithms](https://mlsanity.com/paper/1707.06347/) (2017)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Program Synthesis with Large Language Models](https://mlsanity.com/paper/2108.07732/) (2021)
- [Code Llama: Open Foundation Models for Code](https://mlsanity.com/paper/2308.12950/) (2023)
- [CodeBERT: A Pre-Trained Model for Programming and Natural Languages](https://mlsanity.com/paper/2002.08155/) (2020)
- [Scaling Laws for Reward Model Overoptimization](https://mlsanity.com/paper/2210.10760/) (2022)
- [CodeGen: An Open Large Language Model for Code with Multi-Turn Program Synthesis](https://mlsanity.com/paper/2203.13474/) (2022)
- [CodeBLEU: a Method for Automatic Evaluation of Code Synthesis](https://mlsanity.com/paper/2009.10297/) (2020)
- [Using Human Feedback to Fine-tune Diffusion Models without Any Reward Model](https://mlsanity.com/paper/2311.13231/) (2023)
- [How Abilities in Large Language Models are Affected by Supervised Fine-tuning Data Composition](https://mlsanity.com/paper/2310.05492/) (2023)
- [A Survey of Machine Learning for Big Code and Naturalness](https://mlsanity.com/paper/1709.06182/) (2017)
- [Large Language Models for Code: Security Hardening and Adversarial Testing](https://mlsanity.com/paper/2302.05319/) (2023)
- [Neuro-Symbolic Program Synthesis](https://mlsanity.com/paper/1611.01855/) (2016)
- [SynCoBERT: Syntax-Guided Multi-Modal Contrastive Pre-Training for Code Representation](https://mlsanity.com/paper/2108.04556/) (2021)
- [Self-Supervised Bug Detection and Repair](https://mlsanity.com/paper/2105.12787/) (2021)
- [Retrieval-Based Neural Code Generation](https://mlsanity.com/paper/1808.10025/) (2018)
- [Synthesize, Execute and Debug: Learning to Repair for Neural Program Synthesis](https://mlsanity.com/paper/2007.08095/) (2020)
- [A Case Study of LLM for Automated Vulnerability Repair: Assessing Impact of Reasoning and Patch Validation Feedback](https://mlsanity.com/paper/2405.15690/) (2024)
- [Towards Efficient Fine-tuning of Pre-trained Code Models: An Experimental Study and Beyond](https://mlsanity.com/paper/2304.05216/) (2023)
- [Do Machines and Humans Focus on Similar Code? Exploring Explainability of Large Language Models in Code Summarization](https://mlsanity.com/paper/2402.14182/) (2024)
- [Do Large Language Models Pay Similar Attention Like Human Programmers When Generating Code?](https://mlsanity.com/paper/2306.01220/) (2023)
- [Security Vulnerability Detection Using Deep Learning Natural Language Processing](https://mlsanity.com/paper/2105.02388/) (2021)
- [Modeling Programmer Attention as Scanpath Prediction](https://mlsanity.com/paper/2308.13920/) (2023)
- [CodeGrad: Integrating Multi-Step Verification with Gradient-Based LLM Refinement](https://mlsanity.com/paper/2508.10059/) (2025)
- [Neural Bug Finding: A Study of Opportunities and Challenges](https://mlsanity.com/paper/1906.00307/) (2019)
- [Leveraging Artificial Intelligence on Binary Code Comprehension](https://mlsanity.com/paper/2210.05103/) (2022)
- [DPO-F+: Aligning Code Repair Feedback with Developers' Preferences](https://mlsanity.com/paper/2511.01043/) (2025)

## Built on by (most cited first, 1 of 1)

- [VERITAS: Verifier-Guided Proof Search for Zero-Shot Formal Theorem Proving](https://mlsanity.com/paper/2606.19399/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) can generate plausible code patches, but plausibility is not enough for automated repair: a patch must compile, pass tests, and remove the target vulnerability. We present SynthFix, a neuro-symbolic repair framework that combines supervised repair learning with compiler-informed feedback. During training, a lightweight router selects between Supervised Fine-Tuning (SFT) for common repair patterns and Reward Fine-Tuning (RFT) for examples that benefit from symbolic feedback. The reward combines static structure, lint/compile checks, security scanning, and public execution tests where available; at inference time, the same evidence guides best-of-K candidate selection under a greedy floor. Across five code LLMs (1.3B-7B) on pyrepair, CodeFlaws, and SVEN, SynthFix improves deployable repair metrics over SFT-only and RFT-only baselines, with relative gains up to 54 percent in functional correctness and 14 percent in security clearance. Our code and data are available at https://github.com/CoderDoge1108/SynthFix.