mlsanityData to 14 Oct 2026

Causality-Driven Neural Network Repair: Challenges and Opportunities

arXiv 2504.17946 ↗First posted 24 Apr 2025cs.LGcited by 0

On the map: LLM reasoning → Causal reasoning in LLMs

Abstract (arXiv)

Deep Neural Networks (DNNs) often rely on statistical correlations rather than causal reasoning, limiting their robustness and interpretability. While testing methods can identify failures, effective debugging and repair remain challenging. This paper explores causal inference as an approach primarily for DNN repair, leveraging causal debugging, counterfactual analysis, and structural causal models (SCMs) to identify and correct failures. We discuss in what ways these techniques support fairness, adversarial robustness, and backdoor mitigation by providing targeted interventions. Finally, we discuss key challenges, including scalability, generalization, and computational efficiency, and outline future directions for integrating causality-driven interventions to enhance DNN reliability.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Placed in Causal reasoning in LLMs, one of 17 directions in LLM reasoning. Builds on 13 of its 28 references that are in the map. First posted on arXiv 24 Apr 2025.

Builds on

13 of 28 references.

References · 13

  1. 2019A Survey on Explainable Artificial Intelligence (XAI): Towards Medical XAIcited by 404
  2. 2020Long-Tailed Classification by Keeping the Good and Removing the Bad Momentum Causal Effectcited by 256
  3. 2020Deep Structural Causal Models for Tractable Counterfactual Inferencecited by 211
  4. 2022Revisiting the "Video" in Video-Language Understandingcited by 169
  5. 2020Deep Learning for LiDAR Point Clouds in Autonomous Driving: A Reviewcited by 116
  6. 2021CausalAdv: Adversarial Robustness through the Lens of Causalitycited by 37
  7. 2018Challenges of Using Text Classifiers for Causal Inferencecited by 33
  8. 2019Arachne: Search Based Repair of Deep Neural Networkscited by 26
  9. 2022Causal Reasoning Meets Visual Representation Learning: A Prospective Studycited by 24
  10. 2018Causal Testing: Finding Defects' Root Causescited by 22
  11. 2018Explaining Deep Learning Models using Causal Inferencecited by 18
  12. 2021Causal Learning for Socially Responsible AIcited by 7
1 more
  1. 2023Where and How to Attack? A Causality-Inspired Recipe for Generating Counterfactual Adversarial Examplescited by 3

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2022Causality-based Neural Network Repair
  2. 2024Causal Concept Graph Models: Beyond Causal Opacity in Deep Learning
  3. 2024CausAdv: A Causal-based Framework for Detecting Adversarial Examples
  4. 2025Adversarial Causal Tuning for Realistic Time-series Generation
  5. 2025Provable Repair of Deep Neural Network Defects by Preimage Synthesis and Property Refinement
  6. 2025Causal Deep Q Network
  7. 2026Bucketing the Good Apples: A Method for Diagnosing and Improving Causal Abstraction
  8. 2024CausalBN-Bench: A Comprehensive Benchmark for Causal Learning Capability of LLMs