English(EN) DiaWhisper-DPO: Role-Attributed Transcription of Clinical Interviews via Failure-Mined Preference Optimization

新模型DiaWhisper-DPO改进了临床访谈转录和角色归因

作者 PulseAugur 编辑部 · [1 个来源] · 2026-09-16 04:00

研究人员开发了DiaWhisper-DPO，一个用于转录临床访谈并将话语归因于临床医生或患者的端到端模型。该模型使用LoRA和辅助角色头对Whisper-large-v3进行微调，并通过DiaWhisper-DPO进一步优化性能，该模型利用解码失败作为偏好优化的负面示例，而无需人工标注。该系统在DAIC-WoZ数据集上表现出显著的改进，角色准确率达到0.973，DER比级联基线降低了72%，同时在跨语言PDCH-HAMD数据集上也表现强劲。 AI

影响提高了临床对话分析的准确性，可能改进自动抑郁筛查工具。

排序理由介绍临床访谈转录新模型的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]

在 arXiv cs.CL 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

本文如何被排名

Signal score: 15 / 100; Composite score across the factors below. Higher = stronger signal that this story matters right now.
Newsworthiness bucket: Tool; 介绍临床访谈转录新模型的学术论文。[lever_c_demoted from research: ic=1 ai=1.0]
Source corroboration: Single-source cluster; Only one publisher covered this so far. Single-source stories can still rank when the publisher is high-authority, but they lack cross-source corroboration.
Topics: paper, model release; Editorial topic classification. Feeds into how the story surfaces on /topic/<slug> hub pages and into the per-entity coverage mix.
AI-industry relevance: High; Clearly on-topic for AI-industry coverage.
Story freshness: Breaking (< 6h); Fresh story with cross-source coverage still developing. Ranking may shift as more sources report.

完整方法见我们的编辑标准。

报道来源 [1]

arXiv cs.CL TIER_1 English(EN) · Weiming Li, Ana Catarina Fidalgo Barata, Miguel Constante, Jo\~ao Miguel Sanches · 2026-09-16 04:00

DiaWhisper-DPO：通过故障挖掘偏好优化实现临床访谈的角色归属转录

arXiv:2609.16661v1 Announce Type: new Abstract: 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-synthesizing both sides for con…

本文如何被排名

报道来源 [1]

DiaWhisper-DPO：通过故障挖掘偏好优化实现临床访谈的角色归属转录

相关实体

相关话题