# Data Attribution for Diffusion Models: Timestep-induced Bias in Influence Estimation

arXiv: https://arxiv.org/abs/2401.09031 · HTML: https://mlsanity.com/paper/2401.09031/

## Key facts

- Cited by 14 papers in the map (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 37 of its 44 references that are in the map.
- First posted on arXiv 17 Jan 2024.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 30 of 37)

- [Denoising Diffusion Probabilistic Models](https://mlsanity.com/paper/2006.11239/) (2020)
- [High-Resolution Image Synthesis with Latent Diffusion Models](https://mlsanity.com/paper/2112.10752/) (2021)
- [Denoising Diffusion Implicit Models](https://mlsanity.com/paper/2010.02502/) (2020)
- [Diffusion Models Beat GANs on Image Synthesis](https://mlsanity.com/paper/2105.05233/) (2021)
- [Hierarchical Text-Conditional Image Generation with CLIP Latents](https://mlsanity.com/paper/2204.06125/) (2022)
- [Photorealistic Text-to-Image Diffusion Models with Deep Language Understanding](https://mlsanity.com/paper/2205.11487/) (2022)
- [Classifier-Free Diffusion Guidance](https://mlsanity.com/paper/2207.12598/) (2022)
- [Generative Adversarial Networks](https://mlsanity.com/paper/2203.00667/) (2022)
- [Adding Conditional Control to Text-to-Image Diffusion Models](https://mlsanity.com/paper/2302.05543/) (2023)
- [Deep Learning Face Attributes in the Wild](https://mlsanity.com/paper/1411.7766/) (2014)
- [InstructPix2Pix: Learning to Follow Image Editing Instructions](https://mlsanity.com/paper/2211.09800/) (2022)
- [Prompt-to-Prompt Image Editing with Cross Attention Control](https://mlsanity.com/paper/2208.01626/) (2022)
- [Imagen Video: High Definition Video Generation with Diffusion Models](https://mlsanity.com/paper/2210.02303/) (2022)
- [Understanding Black-box Predictions via Influence Functions](https://mlsanity.com/paper/1703.04730/) (2017)
- [DiffWave: A Versatile Diffusion Model for Audio Synthesis](https://mlsanity.com/paper/2009.09761/) (2020)
- [Palette: Image-to-Image Diffusion Models](https://mlsanity.com/paper/2111.05826/) (2021)
- [Diffusion-LM Improves Controllable Text Generation](https://mlsanity.com/paper/2205.14217/) (2022)
- [Extracting Training Data from Diffusion Models](https://mlsanity.com/paper/2301.13188/) (2023)
- [Estimating Training Data Influence by Tracing Gradient Descent](https://mlsanity.com/paper/2002.08484/) (2020)
- [Does Learning Require Memorization? A Short Tale about a Long Tail](https://mlsanity.com/paper/1906.05271/) (2019)
- [Diffusion Art or Digital Forgery? Investigating Data Replication in Diffusion Models](https://mlsanity.com/paper/2212.03860/) (2022)
- [TRAK: Attributing Model Behavior at Scale](https://mlsanity.com/paper/2303.14186/) (2023)
- [Influence Functions in Deep Learning Are Fragile](https://mlsanity.com/paper/2006.14651/) (2020)
- [Structural Pruning for Diffusion Models](https://mlsanity.com/paper/2305.10924/) (2023)
- [Representer Point Selection for Explaining Deep Neural Networks](https://mlsanity.com/paper/1811.09720/) (2018)
- [Ambient Diffusion: Learning Clean Distributions from Corrupted Data](https://mlsanity.com/paper/2305.19256/) (2023)
- [Influence Selection for Active Learning](https://mlsanity.com/paper/2108.09331/) (2021)
- [The ArtBench Dataset: Benchmarking Generative Models with Artworks](https://mlsanity.com/paper/2206.11404/) (2022)
- [On Memorization in Probabilistic Deep Generative Models](https://mlsanity.com/paper/2106.03216/) (2021)
- [Measuring the Effect of Training Data on Deep Learning Predictions via Randomized Experiments](https://mlsanity.com/paper/2206.10013/) (2022)

## Built on by (most cited first, 14 of 14)

- [A Survey on Mechanistic Interpretability for Multi-Modal Foundation Models](https://mlsanity.com/paper/2502.17516/) (2025)
- [Improving Influence-based Instruction Tuning Data Selection for Balanced Learning of Diverse Capabilities](https://mlsanity.com/paper/2501.12147/) (2025)
- [CUPID: Curating Data your Robot Loves with Influence Functions](https://mlsanity.com/paper/2506.19121/) (2025)
- [A Snapshot of Influence: A Local Data Attribution Framework for Online Reinforcement Learning](https://mlsanity.com/paper/2505.19281/) (2025)
- [Influence-Preserving Proxies for Gradient-Based Data Selection in LLM Fine-tuning](https://mlsanity.com/paper/2602.17835/) (2026)
- [Motion Attribution for Video Generation](https://mlsanity.com/paper/2601.08828/) (2026)
- [Concept-TRAK: Understanding how diffusion models learn concepts through concept-level attribution](https://mlsanity.com/paper/2507.06547/) (2025)
- [Variance Reduction for Expectations with Diffusion Teachers](https://mlsanity.com/paper/2605.21489/) (2026)
- [Enhancing High-Energy Particle Physics Collision Analysis through Graph Data Attribution Techniques](https://mlsanity.com/paper/2407.14859/) (2024)
- [Concept Reachability in Diffusion Models: Beyond Dataset Constraints](https://mlsanity.com/paper/2505.19313/) (2025)
- [Gradient-Aligned Calibration for Post-Training Quantization of Diffusion Models](https://mlsanity.com/paper/2602.01289/) (2026)
- [Tracing Generated Samples to Training-Data Clusters in Flow-Matching Models](https://mlsanity.com/paper/2608.30081/) (2026)
- [Distilling Diffusion Score Discrepancy for Efficient Training Data Attribution](https://mlsanity.com/paper/2609.38776/) (2026)
- [Attributing HOW, Not Just WHICH: Counterfactual Response Trajectories for Diffusion Models](https://mlsanity.com/paper/2610.11238/) (2026)

## Abstract (arXiv)

Data attribution methods trace model behavior back to its training dataset, offering an effective approach to better understand ''black-box'' neural networks. While prior research has established quantifiable links between model output and training data in diverse settings, interpreting diffusion model outputs in relation to training samples remains underexplored. In particular, diffusion models operate over a sequence of timesteps instead of instantaneous input-output relationships in previous contexts, posing a significant challenge to extend existing frameworks to diffusion models directly. Notably, we present Diffusion-TracIn that incorporates this temporal dynamics and observe that samples' loss gradient norms are highly dependent on timestep. This trend leads to a prominent bias in influence estimation, and is particularly noticeable for samples trained on large-norm-inducing timesteps, causing them to be generally influential. To mitigate this effect, we introduce Diffusion-ReTrac as a re-normalized adaptation that enables the retrieval of training samples more targeted to the test sample of interest, facilitating a localized measurement of influence and considerably more intuitive visualization. We demonstrate the efficacy of our approach through various evaluation metrics and auxiliary tasks, reducing the amount of generally influential samples to frac13 of its original quantity.