PaperScope
LIVE · 2026-09-09 05:40 UTC

A Shared-Backbone Approach for Multi-Task MedMNIST Classification

Stefan-Dorian Gavril, Andrei Arhire, Adrian Iftene

Latestcs.CLcs.LGcs.AIcs.CV
arXiv ID
2609.06838 v1
Category
Submitted
2026-09-06

Abstract

Multi-task biomedical classification requires models to generalize across disparate modalities and class distributions. We study 11 heterogeneous MedMNIST datasets using the harmonic mean of per-task macro-F1. We evaluate three backbones with task-specific linear heads. We identify a resolution domain shift between the MedMNIST API and evaluation environment. Resolving this inconsistency and optimizing architecture-specific regularization substantially improved performance. Our best configuration, a ConvNeXt-Tiny backbone with label smoothing, achieved a leaderboard harmonic-mean macro-F1 of 0.73294 in the Tensor Reloaded: Multi-Task MedMNIST competition, ranking sixth at the close of the official competition phase. Our implementation is publicly available at: https://github.com/GavrilStefan-Dorian/A-Shared-Backbone-Approach-for-Multi-Task-MedMNIST-Classification

Comment: Accepted at the International Conference on System Analysis & Intelligent Information Technologies (SAIIT-2026)

arXiv abs page · PDF · same-day batch