让模型自己学着判断训练数据的重要性，提升定位关键样本的准确性。
Learning to Weight Parameters for Training Data Attribution
- 通过数据直接学习参数重要性权重，不依赖人工标注
- 在图像分类、语言建模和扩散模型中均提升定位精度
- 可细粒度区分主题、风格等概念，适合数据溯源与分析
我们研究基于梯度的数据归属问题，旨在识别哪些训练样本最影响某个输出。现有方法或对网络参数一视同仁，或依赖海瑟恩近似推导的隐式权重，未能充分捕捉参数的功能异质性。为此，我们提出一种直接从数据中显式学习参数重要性权重的方法，无需标注标签。该方法在图像分类、语言建模和扩散任务中均提升了归属准确率，并支持对主体、风格等概念的细粒度归属分析。
原文摘要 · Abstract (English)
We study gradient-based data attribution, aiming to identify which training examples most influence a given output. Existing methods for this task either treat network parameters uniformly or rely on implicit weighting derived from Hessian approximations, which do not fully model functional heterogeneity of network parameters. To address this, we propose a method to explicitly learn parameter importance weights directly from data, without requiring annotated labels. Our approach improves attribution accuracy across diverse tasks, including image classification, language modeling, and diffusion, and enables fine-grained attribution for concepts like subject and style.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。