arXiv cs.AI· Changzeng Fu, Ming Shang, Xinyi Zhang, Jiadong Wang, Runhao Zeng, Shiqi Zhao·· 23 小时前AI 评分29
EDA:面向证据落地的多模态抑郁分析框架,先观察、验证再评估
EDA: Observe, Verify, and Assess for Evidence-Grounded Multimodal Depression Analysis
AI 导读
研究提出 EDA 框架,按 Observe、Verify、Assess 三步将长录音中的声学与视觉线索定位为带时间区间和行为测量值的 Evidence Records,并由冻结的多模态智能体核验每条观察是否有源媒体支持。
正文 · 原文
Abstract:Existing multimodal depression assessment methods often summarize long recordings into global behavioral descriptions. This leaves the source and decision relevance of individual observations difficult to inspect. We propose EDA, an evidence-grounded framework organized around Observe, Verify, and Assess. In Observe, literature-informed acoustic and visual cues are localized into source-linked Evidence Records with temporal intervals and behavioral measurements. In Verify, frozen multimodal agents examine whether the cited media support each observation. The simplified Support admission policy retains positively verified records and their review grounds. In Assess, statistical descriptors and temporal events support reference-based behavioral comparison, while PHQ-9-guided text analysis provides complementary symptom evidence. Cross-modal disagreement triggers review of the original text before final selection. Experiments on D-Vlog and LMVD evaluate human-referenced cue agreement and downstream assessment utility separately. AI cue annotations achieve 91.02% accuracy against human consensus labels on shared source segments. With a frozen 7B backbone, EDA Support improves F1 over direct prediction by 7.61 and 9.88 percentage points on D-Vlog Leave-Query-Out and LMVD, respectively. These findings support explicit, source-linked behavioral evidence for inspectable multimodal depression assessment.
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2611.22648 [cs.AI] |
| (or arXiv:2611.22648v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.22648 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Changzeng Fu [view email]
[v1]
Tue, 17 Nov 2026 12:04:34 UTC (916 KB)
来源:arXiv cs.AI · arxiv.org