# Causal-Driven Feature Evaluation for Cross-Domain Image Classification

Source: [Causal-Driven Feature Evaluation for Cross-Domain Image Classification](<https://www.sota2.com/research/paper/697ae97aa2b28e761d87cfaf>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-01-31T05:00:04.715Z

Authors: Chen Cheng, Ang Li

Published: 2026-01-28T02:17:42.000Z

## Abstract

Out-of-distribution (OOD) generalization remains a fundamental challenge in real-world classification, where test distributions often differ substantially from training data. Most existing approaches pursue domain-invariant representations, implicitly assuming that invariance implies reliability. However, features that are invariant across domains are not necessarily causally effective for prediction.   In this work, we revisit OOD classification from a causal perspective and propose to evaluate learned representations based on their necessity and sufficiency under distribution shift. We introduce an explicit segment-level framework that directly measures causal effectiveness across domains, providing a more faithful criterion than invariance alone.   Experiments on multi-domain benchmarks demonstrate consistent improvements in OOD performance, particularly under challenging domain shifts, highlighting the value of causal evaluation for robust generalization.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Domain Generalization on PACS 2→2 generalization](<https://www.sota2.com/research/sota/domain-generalization-on-pacs-2-2-generalization.md>): Domain Generalization; PACS 2→2 domain generalization; Accuracy (A+C -\> P+S): 91.38.

- [Image Classification on Multi-domain MNIST (test)](<https://www.sota2.com/research/sota/image-classification-on-multi-domain-mnist-test.md>): Image Classification; Multi-domain MNIST (test); Acc (Rotated): 95.86.

- [Image Classification on Multi-domain MNIST (train)](<https://www.sota2.com/research/sota/image-classification-on-multi-domain-mnist-train.md>): Image Classification; Multi-domain MNIST (train); Accuracy (Clean): 96.78.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2601.20176>)

[PDF](<https://arxiv.org/pdf/2601.20176.pdf>)