SurrogateSHAP: Training-Free Contributor Attribution for Text-to-Image (T2I) Models

2026-01-29

AI Summary

This work addresses the lack of efficient and fair attribution mechanisms for data contributors in current text-to-image (T2I) generation models, which hinders equitable compensation and sustainable data markets. The authors propose the first attribution framework that operates without retraining: leveraging the inference process of pretrained T2I models, it employs gradient-boosted trees to model a multidimensional utility function—encompassing image quality, aesthetic scores, and diversity—and efficiently approximates Shapley values via an analytical approach. This method substantially reduces computational complexity while accurately identifying key data contributors. In experiments, it successfully pinpoints sources of spurious correlations in clinical images, outperforming existing attribution techniques in both accuracy and efficiency.

Technology Category

• Computer Vision: Diffusion Models for Vision
• Machine Learning: Calibration & Uncertainty Quantification
• Knowledge Representation and Reasoning: Preferences

Application Category

• Economics, Online Markets and Human Computation: Trust and reliance of crowd workers and data experts on GenAI
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

As Text-to-Image (T2I) diffusion models are increasingly used in real-world creative workflows, a principled framework for valuing contributors who provide a collection of data is essential for fair compensation and sustainable data marketplaces. While the Shapley value offers a theoretically grounded approach to attribution, it faces a dual computational bottleneck: (i) the prohibitive cost of exhaustive model retraining for each sampled subset of players (i.e., data contributors) and (ii) the combinatorial number of subsets needed to estimate marginal contributions due to contributor interactions. To this end, we propose SurrogateSHAP, a retraining-free framework that approximates the expensive retraining game through inference from a pretrained model. To further improve efficiency, we employ a gradient-boosted tree to approximate the utility function and derive Shapley values analytically from the tree-based model. We evaluate SurrogateSHAP across three diverse attribution tasks: (i) image quality for DDPM-CFG on CIFAR-20, (ii) aesthetics for Stable Diffusion on Post-Impressionist artworks, and (iii) product diversity for FLUX.1 on Fashion-Product data. Across settings, SurrogateSHAP outperforms prior methods while substantially reducing computational overhead, consistently identifying influential contributors across multiple utility metrics. Finally, we demonstrate that SurrogateSHAP effectively localizes data sources responsible for spurious correlations in clinical images, providing a scalable path toward auditing safety-critical generative models.

Problem

• Text-to-Image
• contributor attribution
• Shapley value
• data valuation
• fair compensation

Innovation

• SurrogateSHAP
• Shapley value
• training-free attribution
• text-to-image models
• data valuation