cs.CVNov 15, 2026

SEAT: Sign-Type Evidence-Audited Transport for Multi-Label Tongue Diagnosis

Authors: Liuyi Xu, Qing Wang, Zikai Liang, An-Yang Lu, Yining Qian

Organizations: College of Information Science and Engineering, Northeastern University, Shenyang, China · College of Sciences, Northeastern University, Shenyang, China · School of Computer and Communication Engineering, Northeastern University at Qinhuangdao, Qinhuangdao 066004, Hebei, China

Abstract

Automated tongue-sign recognition can improve the consistency of non-invasive examinations that otherwise rely on subjective visual judgment. A single tongue image often exhibits multiple coexisting signs characterized by heterogeneous evidence, including localized morphology, global chromaticity, and inter-label dependence. However, existing methods typically rely on shared representations and uniform multi-label heads, making it difficult to verify whether predictions stem from clinically relevant evidence rather than spurious visual correlations. We propose SEAT, an evidence-audited framework that matches distinct evidence types with dedicated prediction and evaluation mechanisms. Built on frozen DINOv2 patch features, SEAT couples mask-guided sign-specific attention for localized morphology, a semi-relaxed optimal transport pathway over LAB/HSV color prototypes to enhance pale-tongue recognition, and a co-occurrence-aware transport regularizer to shape the batch-level predicted label marginal. These components are evaluated through audits matched to their evidence types. On the TongueDx benchmark, SEAT achieves a mean macro-F1 of 72.59 with a standard deviation of 1.04 across six random seeds, with a six-model ensemble reaching 74.15. The ensemble outperforms a backbone-matched baseline and a larger, domain-specialized vision-language model by 8.06 and 2.99 percentage points, respectively. Our results demonstrate that aligning dedicated prediction pathways and audits with sign-specific characteristics improves multi-label accuracy while offering more targeted evidential grounding than uniform post-hoc spatial explanations. Code is available at https://github.com/NEU-TCM-Lab/Sign-type-Evidence-Audited-Transport.

Figures & tables

Explore similar work

CardsList
  1. What Makes Deep Learning Work for Traditional Chinese Medicine Tongue Diagnosis? A Comprehensive Ablation Study

    Jul 30, 2026Longxia Gao, Linan Wang, Yuhe Han +3Medical DiagnosisMedical Image Classification

  2. OmniTongue: A Comprehensive MLLM Benchmark for Automated Tongue Diagnosis in Traditional Chinese Medicine

    Nov 17, 2026Liuyi Xu, Yun Guo, Shusheng Li +3Medical VLMsMedical Diagnosis

  3. BEACON: Distilled Begin/End-Aware Structured Decoding for Continuous Sign Language Segmentation

    Oct 29, 2026Guohou Zhang, Kalin StefanovSign Language RecognitionBoundary-Aware Segmentation