# A Multi-Probe Audit of Clinical-Interview Depression Detection Benchmarks

arXiv: https://arxiv.org/abs/2605.23977 · HTML: https://mlsanity.com/paper/2605.23977/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 36 references that are in the map.
- First posted on arXiv 13 May 2026.

## Builds on (most cited first, 7 of 7)

- [AVEC 2019 Workshop and Challenge: State-of-Mind, Detecting Depression with AI, and Cross-Cultural Affect Recognition](https://mlsanity.com/paper/1907.11510/) (2019)
- [Cross-validation failure: small sample sizes lead to large error bars](https://mlsanity.com/paper/1706.07581/) (2017)
- [MODMA dataset: a Multi-modal Open Dataset for Mental-disorder Analysis](https://mlsanity.com/paper/2002.09283/) (2020)
- [Multi-level Attention network using text, audio and video for Depression Prediction](https://mlsanity.com/paper/1909.01417/) (2019)
- [DAIC-WOZ: On the Validity of Using the Therapist's prompts in Automatic Depression Detection from Clinical Interviews](https://mlsanity.com/paper/2404.14463/) (2024)
- [Who is Speaking or Who is Depressed? A Controlled Study of Speaker Leakage in Speech-Based Depression Detection](https://mlsanity.com/paper/2604.14354/) (2026)
- [When Consistency Becomes Bias: Interviewer Effects in Semi-Structured Clinical Interviews](https://mlsanity.com/paper/2603.24651/) (2026)

## Built on by (most cited first, 1 of 1)

- [Cross-Scale Transfer Learning for Depression Severity Prediction: From PHQ-8 to HAMD-17 Across Languages and Clinical Paradigms](https://mlsanity.com/paper/2609.28430/) (2026)

## Abstract (arXiv)

This paper audits benchmark evaluation in clinical-interview depression detection through four complementary probes across DAIC/E-DAIC, CMDC, ANDROIDS, MODMA, and PDCH. First, we re-evaluate E-DAIC under strict subject-disjoint leave-one-subject-out cross-validation. A lightweight hybrid text-plus-LLM-score model reaches macro-F1 = 0.723 - the highest reported under this protocol, to our knowledge - providing a conservative out-of-fold reference point that does not depend on the privileged official holdout. Second, we test whether the E-DAIC official split supports fine-grained leaderboard rankings by sweeping 96 model configurations across modality bundles, pooling strategies, and learners. Development-side cross-validation and official-test rankings align only moderately: the best cross-validation configuration ranks twentieth on the official test, the official-test winner ranks forty-first by cross-validation, top-3 overlap is zero, and the apparent winner is rank-1 in only 32.3% of subject bootstraps. Third, we externally validate strong public CMDC and ANDROIDS baselines that achieve near-ceiling in-domain performance. Zero-shot transfer to external corpora is substantially weaker. Finally, we stress-test E-DAIC text and audio models using paired symptom-dense versus symptom-light interview slices defined by an SRDS-based annotator. Text scores rise sharply on symptom-dense slices, whereas audio scores remain nearly flat; the text-minus-audio gap is positive across all five seeds.