The Flow has not summarised this story yet — read it at arXiv Machine Learning.
The paper introduces Evidence-Bounded Mental Health Reasoning, addressing the problem that current multimodal mental health screening models treat all clinical speech protocols as equally evidential. It presents the Evidence Package Benchmark, comprising 1,870 annotated packages from six diverse protocols, and proposes EviBound, a protocol-aware framework that limits reasoning to valid evidence using a planner, acoustic consensus, and a boundary critic. EviBound outperforms existing omni-modal baselines, achieving a Depression AUROC of 0.8658 with no claim violations.
The paper introduces DiaWhisper-DPO, an end‑to‑end model that fine‑tunes Whisper-large-v3 with LoRA and a frame‑level role head to transcribe and attribute utterances in clinical interviews. It further refines the system using failure‑mined preference optimization (DPO) that leverages genuine decoding failures as rejected completions, eliminating the need for human preference data. On the DAIC‑WOZ dataset, DiaWhisper‑DPO attains 0.973 role accuracy and 0.119 DER, outperforming cascaded baselines by 72% and dramatically reducing seed variation, while also improving performance on the cross‑lingual PDCH‑HAMD dataset.
arXiv:2608. 19211v1 Announce Type: cross Abstract: Human speech is richly expressive, with prosody carrying linguistic and emotional information beyond the lexical content.
Automated depression screening from clinical interviews requires attribution of utterances to the clinician or patient. We evaluate two datasets: DAIC-WOZ, where participant-only recordings require re...
arXiv:2607. 03744v1 Announce Type: new Abstract: Automatic depression detection from clinical interviews typically models the semantic content and acoustic characteristics of participant speech.
The paper examines automatic depression detection from doctor‑patient conversations and finds that models trained on semi‑structured interview data can achieve high accuracy by exploiting fixed interviewer prompts rather than the participants’ language. Across three datasets (ANDROIDS, DAIC‑WOZ, E‑DAIC), the authors show that restricting models to participant utterances distributes decision evidence more broadly and reflects genuine linguistic cues. The study highlights a cross‑dataset, architecture‑agnostic bias introduced by interviewer prompts and calls for analyses that localize decision evidence by time and speaker to ensure models learn from participants’ language.