Shapley Value Attribution

Ask about this topic

Overview

Shapley value attribution uses cooperative game theory to distribute an outcome among participants according to their average marginal contribution across possible coalitions. It is used to explain model predictions and interactions, value training data, and assign credit in settings such as multimodal models and multi-agent systems. Common approaches include exact computation for structured models, Monte Carlo and regression-based estimators, and hierarchical variants such as Owen values; causal and conditional methods aim to make coalition comparisons more realistic when inputs are dependent. A central challenge is balancing computational cost with reliable explanations, while choices about the value function, missing-feature treatment, and data distribution can substantially change what an attribution means—and whether it helps people make better decisions.

Latest papers

CardsList

103 papers10 papers in the last four weeks, up 233% on the four weeks before. 0.1% of all new papers.

  1. A Polynomial-Time Axiomatic Alternative to SHAP for Feature Attribution

    Feb 28, 2026Kazuhiro Hiraki, Shinichi Ishihara, Takumi Kongo +1Feature AttributionShapley Value Attribution

    International Monetary Fund · Independent Researcher · Fukuoka University +1

  2. Fair feature attribution for multi-output prediction: a Shapley-based perspective

    Feb 26, 2026Umberto Biccari, Alain Ibáñez de Opakua, José María Mato +3Feature AttributionShapley Value Attribution

  3. SGPA: Spectrogram-Guided Phonetic Alignment for Feasible Shapley Value Explanations in Multimodal Large Language Models

    Feb 25, 2026Paweł Pozorski, Jakub Muszyński, Maria GanzhaShapley Value AttributionAudio-Language Models

    Warsaw University of Technology, Warsaw, Poland

  4. cc-Shapley: Measuring Multivariate Feature Importance Needs Causal Context

    Feb 23, 2026Jörg Martin, Stefan HaufeFeature AttributionShapley Value Attribution

    Physikalisch-Technische Bundesanstalt, Berlin, Germany · Technische Universität Berlin, Berlin, Germany · Charité – Universitätsmedizin Berlin, Berlin, Germany

  5. Owen-based Semantics and Hierarchy-Aware Explanation (O-Shap)

    Feb 19, 2026Xiangyu Zhou, Chenhan Xiao, Yang WengFeature AttributionSHAP Feature Attribution

    Department of Electrical, Computer and Energy Engineering, Arizona State University, Tempe, Arizona, United States

  6. Statistical Inference and Learning for Shapley Additive Explanations (SHAP)

    Feb 11, 2026Justin Whitehouse, Ayush Sawarni, Vasilis SyrgkanisSHAP Feature AttributionShapley Value Attribution

    Management Science and Engineering, Stanford University

  7. Shapley-based Data Valuation for LLM Alignment via Sequential Preference Optimization

    Dec 12, 2025Mélissa Tamine, Otmane Sakhi, Benjamin Heymann +2LLM AlignmentShapley Value Attribution

    Criteo AI lab, FairPlay joint team, France · CREST, ENSAE, Institut Polytechnique de Paris · Criteo AI Lab, France +1

  8. Regression-adjusted Monte Carlo Estimators for Shapley Values and Probabilistic Values

    Jun 13, 2025R. Teal Witter, Yurong Liu, Christopher MuscoShapley Value AttributionValue Function Estimation

    Claremont McKenna College · New York University

  9. TRIP-Shapley: Propagation-Aware Contribution Attribution Framework for Peer-to-Peer Federated Learning

    May 29, 2025Honoka Anada, Tatsuya Kaneko, Yafei Ou +1Federated LearningShapley Value Attribution

    The University of Tokyo · Institute of Science Tokyo · RIKEN AIP

  10. Fair Document Valuation in LLM Summaries via Shapley Values

    May 28, 2025Zikun Ye, Hema YoganarasimhanShapley Value AttributionData Valuation

    University of Washington

  11. Provably Accurate Shapley Value Estimation via Leverage Score Sampling

    Oct 2, 2024Christopher Musco, R. Teal WitterSHAP Feature AttributionShapley Value Attribution

    New York University

  12. Explaining a probabilistic prediction on the simplex with Shapley compositions

    Aug 2, 2024Paul-Gauthier Noé, Miquel Perelló-Nieto, Jean-François Bonastre +1Shapley Value Attribution

    Laboratoire Informatique d’Avignon, Avignon Universit´e, France · University of Bristol, United Kingdom

  13. Neural Dynamic Data Valuation via Stochastic State-Adjoint Trajectories

    Apr 30, 2024Zhangyong Liang, Ji Zhang, Huanhuan GaoCooperative Game TheoryGame Theory

    National Center for Applied Mathematics Tianjin University Tianjin 300072, China · School of Mathematics, Physics and Computing University of Southern Queensland Toowoomba, Queensland 4350, Australia · School of Mechanical and Aerospace Engineering Jilin University Changchun 130025, China