MADE: A Living Benchmark for Multi-Label Text Classification with Uncertainty Quantification of Medical Device Adverse Events

TL;DR

MADE is a continuously updated benchmark for multi-label text classification in healthcare, emphasizing uncertainty quantification and addressing challenges like label imbalance and data contamination.

Contribution

It introduces MADE, a living, hierarchical, and contamination-resistant MLTC benchmark with comprehensive evaluation of diverse models and uncertainty methods.

Findings

Smaller fine-tuned decoders achieve high accuracy and competitive UQ.

Generative fine-tuning provides the most reliable uncertainty quantification.

Large reasoning models improve rare label performance but have weak UQ.

Abstract

Machine learning in high-stakes domains such as healthcare requires not only strong predictive performance but also reliable uncertainty quantification (UQ) to support human oversight. Multi-label text classification (MLTC) is a central task in this domain, yet remains challenging due to label imbalances, dependencies, and combinatorial complexity. Existing MLTC benchmarks are increasingly saturated and may be affected by training data contamination, making it difficult to distinguish genuine reasoning capabilities from memorization. We introduce MADE, a living MLTC benchmark derived from {m}edical device {ad}verse {e}vent reports and continuously updated with newly published reports to prevent contamination. MADE features a long-tailed distribution of hierarchical labels and enables reproducible evaluation with strict temporal splits. We establish baselines across more than 20 encoder- and decoder-only models under fine-tuning and few-shot settings (instruction-tuned/reasoning variants, local/API-accessible). We systematically assess entropy-/consistency-based and self-verbalized UQ methods. Results show clear trade-offs: smaller discriminatively fine-tuned decoders achieve the strongest head-to-tail accuracy while maintaining competitive UQ; generative fine-tuning delivers the most reliable UQ; large reasoning models improve performance on rare labels yet exhibit surprisingly weak UQ; and self-verbalized confidence is not a reliable proxy for uncertainty. Our work is publicly available at https://hhi.fraunhofer.de/aml-demonstrator/made-benchmark.