[Short] Real-Time Explanations for Tabular Foundation Models

Luan Borges Teodoro Reis Sena, Francisco Galuppo Azevedo

Keywords: Interpretability; tabular foundation models; Shapley values; SHAP; explanation-aware prediction; scientific machine learning; feature attribution; Prior-Data Fitted Networks

TL;DR: We introduce ShapPFN, a tabular foundation model that outputs predictions and SHAP-style explanations in one forward pass, achieving competitive accuracy with high-fidelity explanations over 1000× faster than KernelSHAP.

Abstract: Interpretability is central for scientific machine learning, as understanding why models make predictions enables hypothesis generation and validation. While tabular foundation models show strong performance, existing explanation methods like SHAP are computationally expensive, limiting interactive exploration. We introduce ShapPFN, a foundation model that integrates Shapley value regression directly into its architecture, producing both predictions and explanations in a single forward pass. On standard benchmarks, ShapPFN achieves competitive performance while producing high-fidelity explanations ($R^2$=0.96, cosine=0.99) over 1000× faster than KernelSHAP (0.06s vs 610s).

Submission Number: 86