Data attribution seeks to trace model outputs back to training data. With the
recent development of diffusion models, data attribution has become a desired
module to properly assign valuations for high-quality or copyrighted training
samples, ensuring that data contributors are fairly compensated or credited.
Several theoretically motivated methods have been proposed to implement data
attribution, in an effort to improve the trade-off between computational
scalability and effectiveness. In this work, we conduct extensive experiments
and ablation studies on attributing diffusion models, specifically focusing on
DDPMs trained on CIFAR-10 and CelebA, as well as a Stable Diffusion model
LoRA-finetuned on ArtBench. Intriguingly, we report counter-intuitive
observations that theoretically unjustified design choices for attribution
empirically outperform previous baselines by a large margin, in terms of both
linear datamodeling score and counterfactual evaluation. Our work presents a
significantly more efficient approach for attributing diffusion models, while
the unexpected findings suggest that at least in non-convex settings,
constructions guided by theoretical assumptions may lead to inferior
attribution performance. The code is available at
this https URL

数据归因通过追溯模型输出至训练数据，以便为高质量或受版权保护的训练样本进行正确的归属评估，确保数据贡献者得到公正的补偿或认可。本研究在扩散模型上进行了广泛的实验和消融研究，特别关注 DDPMs 在 CIFAR-10、CelebA 以及在 ArtBench 上经过 LoRA 微调的稳定扩散模型的归因问题。有趣的是，我们报告了一些违背直觉的观察结果，即在理论上没有依据的设计选择在实践中表现出比以往基准线更好的性能，无论是线性数据建模得分还是对事实的评估。我们提出了一种更高效的方法来为扩散模型进行归因，而意外的发现表明，至少在非凸设置中，受到理论假设指导的构建可能导致较差的归因性能。该文提供了代码链接供参考。

扩散模型数据归因的有趣特性

Intriguing Properties of Data Attribution on Diffusion Models

We provide simple and approximately revenue-optimal mechanisms in the
multi-item multi-bidder settings. We unify and improve all previous results, as
well as generalize the results to broader cases. In particular, we prove that
the better of the following two simple, deterministic and Dominant Strategy
Incentive Compatible mechanisms, a sequential posted price mechanism or an
anonymous sequential posted price mechanism with entry fee, achieves a constant
fraction of the optimal revenue among all randomized, Bayesian Incentive
Compatible mechanisms, when buyers' valuations are XOS over independent items.
If the buyers' valuations are subadditive over independent items, the
approximation factor degrades to $O(\log m)$, where $m$ is the number of items.
We obtain our results by first extending the Cai-Devanur-Weinberg duality
framework to derive an effective benchmark of the optimal revenue for
subadditive bidders, and then analyzing this upper bound with new techniques.

该论文提出了在多个项目和多个竞标人设置中的简单又近似最优机制，其中分别提到了简单的确定和支配策略激励兼备的顺序发布价格机制或带进场费的匿名顺序发布价格机制。

通过对偶理论实现次可加买家的简单机制

Simple Mechanisms for Subadditive Buyers via Duality

In most of microeconomic theory, consumers are assumed to exhibit decreasing
marginal utilities. This paper considers combinatorial auctions among such
submodular buyers. The valuations of such buyers are placed within a hierarchy
of valuations that exhibit no complementarities, a hierarchy that includes also
OR and XOR combinations of singleton valuations, and valuations satisfying the
gross substitutes property. Those last valuations are shown to form a
zero-measure subset of the submodular valuations that have positive measure.
While we show that the allocation problem among submodular valuations is
NP-hard, we present an efficient greedy 2-approximation algorithm for this case
and generalize it to the case of limited complementarities. No such
approximation algorithm exists in a setting allowing for arbitrary
complementarities. Some results about strategic aspects of combinatorial
auctions among players with decreasing marginal utilities are also presented.

本研究考虑了降低边际效用的买家之间的组合拍卖问题，使用了层次化的价格结构，采用了贪心算法来近似求解 NP-hard 问题。