Training data attribution in diffusion models via mirrored unlearning and noise-consistent skew

arXiv:2605.17938 · cs.LG, cs.AI, stat.ML · Submitted 2026-05-18 · Read on arXiv

cs.LG, cs.AI, stat.ML

Submitted: 2026-05-18

Updated: 2026-10-13

Terminology

• Diffusion Model — 2× in this paper · explained in 9 episodes
• Diffusion Models — 2× in this paper · explained in 6 episodes
• Core — 1× in this paper · explained in 2 episodes
• Downstream Tasks — 1× in this paper · explained in 2 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Interpretability — 1× in this paper · explained in 15 episodes
• Robustness — 1× in this paper · explained in 5 episodes

Sources

• Attribution-by-design: Ensuring Inference-Time Provenance in Generative Music Systems
• The Principles of Diffusion Models
• Nonparametric Data Attribution for Diffusion Models
• Ablation Based Counterfactuals
• Distributional Training Data Attribution: What do Influence Functions Sample?
• Fast Data Attribution for Text-to-Image Models
• ReTrack: Data Unlearning in Diffusion Models through Redirecting the Denoising Trajectory
• LLM Surgery: Efficient Knowledge Unlearning and Editing in Large Language Models
• The ArtBench Dataset: Benchmarking Generative Models with Artworks
• Revisiting Methods for Finding Influential Examples

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks