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 $\frac{1}{3}$ of its original quantity.
SIPO: Stabilized and Improved Preference Optimization for Aligning Diffusion Models
Shanghai Science and Intelligence Institute, Shanghai, China(上海科学与智能研究所);Fudan University, Shanghai, China(复旦大学);Australian Institute for Machine Learning, The University of Adelaide(澳大利亚机器学习研究所,阿德莱德大学)
BiasMap: Leveraging Cross-Attentions to Discover and Mitigate Hidden Social Biases in Text-to-Image Generation
Institute for Clarity in Documentation(清晰文档研究所);Inria Paris-Rocquencourt(巴黎-罗克琴court研究所);Rajiv Gandhi University(拉吉夫·甘地大学);Tsinghua University(清华大学);Palmer Research Laboratories(帕勒尔研究实验室);University of North Carolina at Charlotte(北卡罗来纳大学教堂山分校);Utah State University(犹他州立大学)
Fraunhofer IPA(弗劳恩霍夫生产技术与自动化研究所);CNR Institute of Applied Sciences and Intelligent Systems(意大利国家研究委员会应用科学与智能系统研究所);IDS Imaging Development Systems GmbH(IDS成像开发系统有限公司);University of Stuttgart(斯图加特大学)
Revealing Hidden Context Bias in Segmentation and Object Detection through Concept-specific Explanations
Fraunhofer Heinrich-Hertz-Institute(弗劳恩霍夫海因里希·赫兹研究所);Technische Universität Berlin(柏林工业大学);Berlin Institute for the Foundations of Learning and Data(柏林学习与数据基础研究所)
What's Pulling the Strings? Evaluating Integrity and Attribution in AI Training and Inference through Concept Shift
University of New South Wales \& CSIRO's Data61 Sydney Australia;Macquarie University Sydney Australia;University of New South Wales Sydney Australia;University of Illinois at Urbana–Champaign Champaign;University of New South Wales \& CSIRO's Data61;Macquarie University;University of New South Wales;University of Illinois at Urbana–Champaign
Mohamed bin Zayed University of Artificial Intelligence(莫萨德·本·泽德人工智能大学);University at Buffalo(布法罗大学);King Abdullah University of Science and Technology(卡布斯国王科学与技术大学)