mlsanityData to 14 Oct 2026

On the Fragility of Data Attribution When Learning Is Distributed

arXiv 2605.15520 ↗First posted 15 May 2026cs.LG cs.AI cs.DCcited by 0

Not on the map. Why?

Abstract (arXiv)

Data attribution has become an important component of pricing, auditing, and governance in machine learning pipelines, yet most attribution methods implicitly assume that attribution values faithfully reflect participants' contributions. We show that this assumption can fail: a single participant in a standard distributed training workflow can substantially inflate its measured attribution value while preserving global utility. Our attribution-first attack uses latent optimization to inject small synthetic batches that preserve utility while exploiting non-IID label coverage and evaluator sensitivities. Across datasets, models, and multiple marginal-utility evaluators, the attack consistently increases the adversary's attribution value and reshapes the relative attribution structure among benign clients without degrading accuracy or triggering geometry-based defenses. These results show that attribution itself forms a new attack surface and motivate the development of attribution-robust and incentive-compatible scoring mechanisms.

Key facts. No citing papers in the map yet (data through 14 Oct 2026). Builds on 57 of its 80 references that are in the map. First posted on arXiv 15 May 2026.

Builds on

57 of 80 references.

References · 57

  1. 2015Deep Residual Learning for Image Recognitioncited by 52,096
  2. 2014Very Deep Convolutional Networks for Large-Scale Image Recognitioncited by 15,364
  3. 2016Communication-Efficient Learning of Deep Networks from Decentralized Datacited by 7,669
  4. 2017Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithmscited by 4,349
  5. 2016Wide Residual Networkscited by 3,070
  6. 2018Federated Learning with Non-IID Datacited by 1,108
  7. 2017Certified Defenses for Data Poisoning Attackscited by 303
  8. 2020Threats to Federated Learning: A Surveycited by 178
  9. 2022Training Data Influence Analysis and Estimation: A Surveycited by 135
  10. 2020A Principled Approach to Data Valuation for Federated Learningcited by 127
  11. 2023DataInf: Efficiently Estimating Data Influence in LoRA-tuned LLMs and Diffusion Modelscited by 123
  12. 2015Practical Bayesian model evaluation using leave-one-out cross-validation and WAICcited by 110
38 more
  1. 2024Data Shapley in One Training Runcited by 81
  2. 2020Free-rider Attacks on Model Aggregation in Federated Learningcited by 57
  3. 2023OpenDataVal: a Unified Benchmark for Data Valuationcited by 55
  4. 2023Intriguing Properties of Data Attribution on Diffusion Modelscited by 47
  5. 2023Recent Advances on Federated Learning: A Systematic Surveycited by 41
  6. 2024Machine Unlearning Fails to Remove Data Poisoning Attackscited by 37
  7. 2024Influence Functions for Scalable Data Attribution in Diffusion Modelscited by 33
  8. 2024Rethinking Data Shapley for Data Selection Tasks: Misleads and Meritscited by 29
  9. 2024Data Attribution for Text-to-Image Models by Unlearning Synthesized Imagescited by 22
  10. 2024Diffusion Attribution Score: Evaluating Training Data Influence in Diffusion Modelscited by 22
  11. 2024Stochastic Amortization: A Unified Approach to Accelerate Feature and Data Attributioncited by 21
  12. 2025Data Poisoning in Deep Learning: A Surveycited by 20
  13. 2022Fair and efficient contribution valuation for vertical federated learningcited by 18
  14. 2024Redefining Contributions: Shapley-Driven Federated Learningcited by 17
  15. 2025Position: The Most Expensive Part of an LLM should be its Training Datacited by 16
  16. 2024Threats and Defenses in Federated Learning Life Cycle: A Comprehensive Survey and Challengescited by 15
  17. 2024Helpful or Harmful Data? Fine-tuning-free Shapley Attribution for Explaining Language Model Predictionscited by 14
  18. 2023DU-Shapley: A Shapley Value Proxy for Efficient Dataset Valuationcited by 12
  19. 2024Enhancing Post-Hoc Attributions in Long Document Comprehension via Coarse Grained Answer Decompositioncited by 12
  20. 2023A Bayesian Approach To Analysing Training Data Attribution In Deep Learningcited by 11
  21. 2025A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learningcited by 11
  22. 2024Efficient Sketches for Training Data Attribution and Studying the Loss Landscapecited by 10
  23. 2024Scaling Laws for the Value of Individual Data Points in Machine Learningcited by 10
  24. 2025Better Training Data Attribution via Better Inverse Hessian-Vector Productscited by 10
  25. 2024Benchmarking the Attribution Quality of Vision Modelscited by 8
  26. 2025Fairshare Data Pricing via Data Valuation for Large Language Modelscited by 8
  27. 2025How to Probe: Simple Yet Effective Techniques for Improving Post-hoc Explanationscited by 8
  28. 2025Enhancing Training Data Attribution with Representational Optimizationcited by 8
  29. 20232D-Shapley: A Framework for Fragmented Data Valuationcited by 7
  30. 2025Fast Data Attribution for Text-to-Image Modelscited by 7
  31. 2024Federated Learning Under Attack: Exposing Vulnerabilities through Data Poisoning Attacks in Computer Networkscited by 6
  32. 2024Tracing Back the Malicious Clients in Poisoning Attacks to Federated Learningcited by 6
  33. 2025Adversarial Training for Defense Against Label Poisoning Attackscited by 6
  34. 2024Data Distribution Valuationcited by 5
  35. 2024A Versatile Influence Function for Data Attribution with Non-Decomposable Losscited by 5
  36. 2024Final-Model-Only Data Attribution with a Unifying View of Gradient-Based Methodscited by 5
  37. 2024An Efficient Framework for Crediting Data Contributors of Diffusion Modelscited by 4
  38. 20242D-OOB: Attributing Data Contribution Through Joint Valuation Frameworkcited by 4

50 most cited of 57 All 57 (JSON)

Built on by

Not cited yet.