# Semantic Code Repair using Neuro-Symbolic Transformation Networks

arXiv: https://arxiv.org/abs/1710.11054 · HTML: https://mlsanity.com/paper/1710.11054/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 30 Oct 2017.

## Built on by (most cited first, 13 of 13)

- [Unifying the Perspectives of NLP and Software Engineering: A Survey on Language Models for Code](https://mlsanity.com/paper/2311.07989/) (2023)
- [Generating Bug-Fixes Using Pretrained Transformers](https://mlsanity.com/paper/2104.07896/) (2021)
- [DeepDebug: Fixing Python Bugs Using Stack Traces, Backtranslation, and Code Skeletons](https://mlsanity.com/paper/2105.09352/) (2021)
- [Neuro-Symbolic AI: Explainability, Challenges, and Future Trends](https://mlsanity.com/paper/2411.04383/) (2024)
- [DeepCode AI Fix: Fixing Security Vulnerabilities with Large Language Models](https://mlsanity.com/paper/2402.13291/) (2024)
- [Learning to Make Compiler Optimizations More Effective](https://mlsanity.com/paper/2102.13514/) (2021)
- [Neural Machine Translation for Code Generation](https://mlsanity.com/paper/2305.13504/) (2023)
- [AuPair: Golden Example Pairs for Code Repair](https://mlsanity.com/paper/2502.18487/) (2025)
- [LAMNER: Code Comment Generation Using Character Language Model and Named Entity Recognition](https://mlsanity.com/paper/2204.09654/) (2022)
- [Example-based Synthesis of Static Analysis Rules](https://mlsanity.com/paper/2204.08643/) (2022)
- [CONCORD: Towards a DSL for Configurable Graph Code Representation](https://mlsanity.com/paper/2401.17967/) (2024)
- [Bridging the Gap: Representation Spaces in Neuro-Symbolic AI](https://mlsanity.com/paper/2411.04393/) (2024)
- [Toward Neurosymbolic Program Comprehension](https://mlsanity.com/paper/2502.01806/) (2025)

## Abstract (arXiv)

We study the problem of semantic code repair, which can be broadly defined as automatically fixing non-syntactic bugs in source code. The majority of past work in semantic code repair assumed access to unit tests against which candidate repairs could be validated. In contrast, the goal here is to develop a strong statistical model to accurately predict both bug locations and exact fixes without access to information about the intended correct behavior of the program. Achieving such a goal requires a robust contextual repair model, which we train on a large corpus of real-world source code that has been augmented with synthetically injected bugs. Our framework adopts a two-stage approach where first a large set of repair candidates are generated by rule-based processors, and then these candidates are scored by a statistical model using a novel neural network architecture which we refer to as Share, Specialize, and Compete. Specifically, the architecture (1) generates a shared encoding of the source code using an RNN over the abstract syntax tree, (2) scores each candidate repair using specialized network modules, and (3) then normalizes these scores together so they can compete against one another in comparable probability space. We evaluate our model on a real-world test set gathered from GitHub containing four common categories of bugs. Our model is able to predict the exact correct repair 41% of the time with a single guess, compared to 13% accuracy for an attentional sequence-to-sequence model.