# Learning to Fix Build Errors with Graph2Diff Neural Networks

arXiv: https://arxiv.org/abs/1911.01205 · HTML: https://mlsanity.com/paper/1911.01205/

## Key facts

- Cited by 26 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 4 Nov 2019.

## Built on by (most cited first, 26 of 26)

- [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)
- [RepairAgent: An Autonomous, LLM-Based Agent for Program Repair](https://mlsanity.com/paper/2403.17134/) (2024)
- [Leveraging Automated Unit Tests for Unsupervised Code Translation](https://mlsanity.com/paper/2110.06773/) (2021)
- [Break-It-Fix-It: Unsupervised Learning for Program Repair](https://mlsanity.com/paper/2106.06600/) (2021)
- [DOBF: A Deobfuscation Pre-Training Objective for Programming Languages](https://mlsanity.com/paper/2102.07492/) (2021)
- [Self-Supervised Bug Detection and Repair](https://mlsanity.com/paper/2105.12787/) (2021)
- [A Syntax-Guided Edit Decoder for Neural Program Repair](https://mlsanity.com/paper/2106.08253/) (2021)
- [CoditT5: Pretraining for Source Code and Natural Language Editing](https://mlsanity.com/paper/2208.05446/) (2022)
- [Neural Transfer Learning for Repairing Security Vulnerabilities in C Code](https://mlsanity.com/paper/2104.08308/) (2021)
- [InstructCoder: Instruction Tuning Large Language Models for Code Editing](https://mlsanity.com/paper/2310.20329/) (2023)
- [On Multi-Modal Learning of Editing Source Code](https://mlsanity.com/paper/2108.06645/) (2021)
- [Deep Learning for Code Intelligence: Survey, Benchmark and Toolkit](https://mlsanity.com/paper/2401.00288/) (2023)
- [VELVET: a noVel Ensemble Learning approach to automatically locate VulnErable sTatements](https://mlsanity.com/paper/2112.10893/) (2021)
- [Learning Structural Edits via Incremental Tree Transformations](https://mlsanity.com/paper/2101.12087/) (2021)
- [KNOD: Domain Knowledge Distilled Tree Decoder for Automated Program Repair](https://mlsanity.com/paper/2302.01857/) (2023)
- [Fix Bugs with Transformer through a Neural-Symbolic Edit Grammar](https://mlsanity.com/paper/2204.06643/) (2022)
- [LExecutor: Learning-Guided Execution](https://mlsanity.com/paper/2302.02343/) (2023)
- [ZeroLeak: Using LLMs for Scalable and Cost Effective Side-Channel Patching](https://mlsanity.com/paper/2308.13062/) (2023)
- [A Survey of Deep Learning Models for Structural Code Understanding](https://mlsanity.com/paper/2205.01293/) (2022)
- [A Library for Representing Python Programs as Graphs for Machine Learning](https://mlsanity.com/paper/2208.07461/) (2022)
- [Learning to Extend Program Graphs to Work-in-Progress Code](https://mlsanity.com/paper/2105.14038/) (2021)
- [AoA: Theorem Proving Agent over Abstract Syntax Tree of Redesigned Language](https://mlsanity.com/paper/2607.16372/) (2026)
- [Graph Neural Networks For Mapping Variables Between Programs -- Extended Version](https://mlsanity.com/paper/2307.13014/) (2023)
- [Evaluating LLM Agents on Automated Software Analysis Tasks](https://mlsanity.com/paper/2604.11270/) (2026)
- [Agentic Generation of AST Transformation Rules for Fixing Breaking Updates](https://mlsanity.com/paper/2606.24446/) (2026)

## Abstract (arXiv)

Professional software developers spend a significant amount of time fixing builds, but this has received little attention as a problem in automatic program repair. We present a new deep learning architecture, called Graph2Diff, for automatically localizing and fixing build errors. We represent source code, build configuration files, and compiler diagnostic messages as a graph, and then use a Graph Neural Network model to predict a diff. A diff specifies how to modify the code's abstract syntax tree, represented in the neural network as a sequence of tokens and of pointers to code locations. Our network is an instance of a more general abstraction that we call Graph2Tocopo, which is potentially useful in any development tool for predicting source code changes. We evaluate the model on a dataset of over 500k real build errors and their resolutions from professional developers. Compared to the approach of DeepDelta (Mesbah et al., 2019), our approach tackles the harder task of predicting a more precise diff but still achieves over double the accuracy.