Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Diffusion Integrated Gradients: Controllable Path Generation for Flexible Feature Attribution

About

Path-based attribution methods such as Integrated Gradients (IG) are widely adopted for their strong axiomatic properties and effectiveness in attributing model predictions to input features by integrating gradients along a path from a baseline to the input. However, the choice of the attribution path largely affects the quality of explanations, and existing approaches rely on fixed or hand-crafted paths that often produce noisy or distorted attributions. To address this limitation, we propose Diffusion Integrated Gradients (DiffIG), a novel method that reformulates path generation as a conditional generative modeling problem. DiffIG first trains a diffusion model to learn a distribution over paths generated from a Stick-Breaking Process, then employs guided sampling to embed user guidance during the sampling procedure. We demonstrate that DiffIG quantitatively matches or outperforms existing path-based methods, achieving perceptually aligned explanations. This work introduces a new generative perspective for flexible, inference-time controllable Explainable Artificial Intelligence (XAI) methods.

Soyeon Kim, Kyowoon Lee, Jaesik Choi• 2026

Related benchmarks

TaskDatasetResultRank
Feature AttributionOxford-IIIT Pet full (val)
DiffID63.68
24
Feature AttributionMini-ImageNet 500 randomly sampled images (val)
DiffID Score42.98
24
Attribution FaithfulnessOxford-IIIT Pet (test)
DiffID57.11
8
Model AttributionMini-ImageNet
DiffID0.4298
6
Showing 4 of 4 rows

Other info

Follow for update