# Causality-based Neural Network Repair

arXiv: https://arxiv.org/abs/2204.09274 · HTML: https://mlsanity.com/paper/2204.09274/

## Key facts

- Cited by 47 papers in the map (data through 14 Oct 2026).
- Placed in Backdoor and data poisoning attacks, one of 12 directions in LLM safety and security.
- Builds on 32 of its 70 references that are in the map.
- First posted on arXiv 20 Apr 2022.

## Placed in

- LLM safety and security / [Backdoor and data poisoning attacks](https://mlsanity.com/safety/backdoor-and-data-poisoning-attacks/)

## Builds on (most cited first, 30 of 32)

- [Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms](https://mlsanity.com/paper/1708.07747/) (2017)
- [FaceNet: A Unified Embedding for Face Recognition and Clustering](https://mlsanity.com/paper/1503.03832/) (2015)
- [End to End Learning for Self-Driving Cars](https://mlsanity.com/paper/1604.07316/) (2016)
- [Counterfactual Fairness](https://mlsanity.com/paper/1703.06856/) (2017)
- [Reluplex: An Efficient SMT Solver for Verifying Deep Neural Networks](https://mlsanity.com/paper/1702.01135/) (2017)
- [Certifying and removing disparate impact](https://mlsanity.com/paper/1412.3756/) (2014)
- [Formal Verification of Piece-Wise Linear Feed-Forward Neural Networks](https://mlsanity.com/paper/1705.01320/) (2017)
- [Formal Security Analysis of Neural Networks using Symbolic Intervals](https://mlsanity.com/paper/1804.10829/) (2018)
- [CausalGAN: Learning Causal Implicit Generative Models with Adversarial Training](https://mlsanity.com/paper/1709.02023/) (2017)
- [Counterfactual Fairness in Text Classification through Robustness](https://mlsanity.com/paper/1809.10610/) (2018)
- [Branch and Bound for Piecewise Linear Neural Network Verification](https://mlsanity.com/paper/1909.06588/) (2019)
- [Automated Directed Fairness Testing](https://mlsanity.com/paper/1807.00468/) (2018)
- [Neural Network Attributions: A Causal Perspective](https://mlsanity.com/paper/1902.02302/) (2019)
- [An Abstraction-Based Framework for Neural Network Verification](https://mlsanity.com/paper/1910.14574/) (2019)
- [DeepMutation: Mutation Testing of Deep Learning Systems](https://mlsanity.com/paper/1805.05206/) (2018)
- [Deep Neural Network Compression for Aircraft Collision Avoidance Systems](https://mlsanity.com/paper/1810.04240/) (2018)
- [Verifying Individual Fairness in Machine Learning Models](https://mlsanity.com/paper/2006.11737/) (2020)
- [Probabilistic Verification of Fairness Properties via Concentration](https://mlsanity.com/paper/1812.02573/) (2018)
- [Provable Repair of Deep Neural Networks](https://mlsanity.com/paper/2104.04413/) (2021)
- [POPQORN: Quantifying Robustness of Recurrent Neural Networks](https://mlsanity.com/paper/1905.07387/) (2019)
- [Verifying Recurrent Neural Networks using Invariant Inference](https://mlsanity.com/paper/2004.02462/) (2020)
- [NNrepair: Constraint-based Repair of Neural Network Classifiers](https://mlsanity.com/paper/2103.12535/) (2021)
- [Causal Learning and Explanation of Deep Neural Networks via Autoencoded Activations](https://mlsanity.com/paper/1802.00541/) (2018)
- [Causal Testing: Finding Defects' Root Causes](https://mlsanity.com/paper/1809.06991/) (2018)
- [Automated Test Generation to Detect Individual Discrimination in AI Models](https://mlsanity.com/paper/1809.03260/) (2018)
- [Explaining Deep Learning Models using Causal Inference](https://mlsanity.com/paper/1811.04376/) (2018)
- [Explaining Visual Models by Causal Attribution](https://mlsanity.com/paper/1909.08891/) (2019)
- [Probabilistic Verification of Neural Networks Against Group Fairness](https://mlsanity.com/paper/2107.08362/) (2021)
- [Towards Repairing Neural Networks Correctly](https://mlsanity.com/paper/2012.01872/) (2020)
- [SOCRATES: Towards a Unified Platform for Neural Network Analysis](https://mlsanity.com/paper/2007.11206/) (2020)

## Built on by (most cited first, 30 of 47)

- [Adaptive Fairness Improvement Based on Causality Analysis](https://mlsanity.com/paper/2209.07190/) (2022)
- [Eliminating Information Leakage in Hard Concept Bottleneck Models with Supervised, Hierarchical Concept Learning](https://mlsanity.com/paper/2402.05945/) (2024)
- [Benchmarking and Explaining Large Language Model-based Code Generation: A Causality-Centric Approach](https://mlsanity.com/paper/2310.06680/) (2023)
- [LLMScan: Causal Scan for LLM Misbehavior Detection](https://mlsanity.com/paper/2410.16638/) (2024)
- [Applications of statistical causal inference in software engineering](https://mlsanity.com/paper/2211.11482/) (2022)
- [Causality Analysis for Evaluating the Security of Large Language Models](https://mlsanity.com/paper/2312.07876/) (2023)
- [NeuFair: Neural Network Fairness Repair with Dropout](https://mlsanity.com/paper/2407.04268/) (2024)
- [Machine Learning Robustness: A Primer](https://mlsanity.com/paper/2404.00897/) (2024)
- [Causality-Aided Trade-off Analysis for Machine Learning Fairness](https://mlsanity.com/paper/2305.13057/) (2023)
- [Online Safety Analysis for LLMs: a Benchmark, an Assessment, and a Path Forward](https://mlsanity.com/paper/2404.08517/) (2024)
- [Neuron Patching: Semantic-based Neuron-level Language Model Repair for Code Generation](https://mlsanity.com/paper/2312.05356/) (2023)
- [Certified Continual Learning for Neural Network Regression](https://mlsanity.com/paper/2407.06697/) (2024)
- [Repairing Deep Neural Networks Based on Behavior Imitation](https://mlsanity.com/paper/2305.03365/) (2023)
- [LUNA: A Model-Based Universal Analysis Framework for Large Language Models](https://mlsanity.com/paper/2310.14211/) (2023)
- [Semantic-Based Neural Network Repair](https://mlsanity.com/paper/2306.07995/) (2023)
- [How Generalizable are Deepfake Image Detectors? An Empirical Study](https://mlsanity.com/paper/2308.04177/) (2023)
- [Causal Reasoning in Software Quality Assurance: A Systematic Review](https://mlsanity.com/paper/2408.17183/) (2024)
- [Causal-Guided Detoxify Backdoor Attack of Open-Weight LoRA Models](https://mlsanity.com/paper/2512.19297/) (2025)
- [Incremental Satisfiability Modulo Theory for Verification of Deep Neural Networks](https://mlsanity.com/paper/2302.06455/) (2023)
- [FITNESS: A Causal De-correlation Approach for Mitigating Bias in Machine Learning Software](https://mlsanity.com/paper/2305.14396/) (2023)
- [Enabling Runtime Verification of Causal Discovery Algorithms with Automated Conditional Independence Reasoning (Extended Version)](https://mlsanity.com/paper/2309.05264/) (2023)
- [A Semantic-based Optimization Approach for Repairing LLMs: Case Study on Code Generation](https://mlsanity.com/paper/2503.12899/) (2025)
- [A Comprehensive Study of Deep Learning Model Fixing Approaches](https://mlsanity.com/paper/2512.23745/) (2025)
- [Reasoning-Based Software Testing](https://mlsanity.com/paper/2303.01302/) (2023)
- [Towards Better Fairness-Utility Trade-off: A Comprehensive Measurement-Based Reinforcement Learning Framework](https://mlsanity.com/paper/2307.11379/) (2023)
- [TraceFL: Interpretability-Driven Debugging in Federated Learning via Neuron Provenance](https://mlsanity.com/paper/2312.13632/) (2023)
- [Inference-Time Rule Eraser: Fair Recognition via Distilling and Removing Biased Rules](https://mlsanity.com/paper/2404.04814/) (2024)
- [NeuSemSlice: Towards Effective DNN Model Maintenance via Neuron-level Semantic Slicing](https://mlsanity.com/paper/2407.20281/) (2024)
- [NeuroBreak: Unveil Internal Jailbreak Mechanisms in Large Language Models](https://mlsanity.com/paper/2509.03985/) (2025)
- [AtPatch: Debugging Transformers via Hot-Fixing Over-Attention](https://mlsanity.com/paper/2601.21695/) (2026)

## Abstract (arXiv)

Neural networks have had discernible achievements in a wide range of applications. The wide-spread adoption also raises the concern of their dependability and reliability. Similar to traditional decision-making programs, neural networks can have defects that need to be repaired. The defects may cause unsafe behaviors, raise security concerns or unjust societal impacts. In this work, we address the problem of repairing a neural network for desirable properties such as fairness and the absence of backdoor. The goal is to construct a neural network that satisfies the property by (minimally) adjusting the given neural network's parameters (i.e., weights). Specifically, we propose CARE (CAusality-based REpair), a causality-based neural network repair technique that 1) performs causality-based fault localization to identify the `guilty' neurons and 2) optimizes the parameters of the identified neurons to reduce the misbehavior. We have empirically evaluated CARE on various tasks such as backdoor removal, neural network repair for fairness and safety properties. Our experiment results show that CARE is able to repair all neural networks efficiently and effectively. For fairness repair tasks, CARE successfully improves fairness by 61.91% on average. For backdoor removal tasks, CARE reduces the attack success rate from over 98% to less than 1%. For safety property repair tasks, CARE reduces the property violation rate to less than 1%. Results also show that thanks to the causality-based fault localization, CARE's repair focuses on the misbehavior and preserves the accuracy of the neural networks.