cs.CLNov 14, 2026

On Representational and Functional Alignment of Training Data Attribution Methods for Language Models

Authors: Loris Schoenegger, Benjamin Roth

Organizations: Faculty of Computer Science, University of Vienna, Vienna, Austria · UniVie Doctoral School Computer Science, University of Vienna, Vienna, Austria · Faculty of Philological and Cultural Studies, University of Vienna, Vienna, Austria

Abstract

Existing approaches to training data attribution derive influence estimates from different sources of information, including gradients, model-internal states, and surface-level similarity. However, comparisons between attribution methods remain incomplete, often relying on indirect evaluations through downstream task performance. In this work, we introduce a framework for directly comparing methods along three dimensions: functional alignment, which measures correlation between full attribution vectors; representational alignment, which captures structural similarity of the representations underlying attribution; and cross-method predictability, which evaluates agreement on local pairwise ranking decisions. We observe strong representational and moderate functional alignment within paradigms. Alignment across paradigms is low and task-dependent, yet stable across models. Our results suggest that the similar downstream performance observed across paradigms in prior work can be attributed to agreement on the relative importance of documents, rather than to agreement on rankings of the full training data.

Figures & tables

Appendix figures & tables18 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

CardsList
  1. Elucidating the Design Space of LM Data Attribution

    Oct 29, 2026Steven Yu, Sheng-Yu Wang, Kimia Hamidieh +2Training Data AttributionGradient-Based Attribution

  2. Measuring Task-Agnostic Training Data Influence Across Language Model Pretraining

    Aug 13, 2026Yuto Nishida, Hirokazu Kiyomaru, Yusuke Oda +6Language Model PretrainingGradient-Based Attribution