Metraipolis › archive › dc_e73b4fda0414
Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols — arXiv:2608.31014
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.31014
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-17T08:07:51.998Z
- SHA-256 of the archived text
e4bb4b791713ad9e1c0d727cd8e65478e4c8392f1d2b4e994e48029cdb54bfb0- Stable URL of this page
https://metraipolis.com/documento/dc_e73b4fda0414
How to cite
Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols — arXiv:2608.31014 — arXiv (en), https://arxiv.org/abs/2608.31014 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_e73b4fda0414 (Metraipolis archive dc_e73b4fda0414, sha256 e4bb4b791713ad9e1c0d727cd8e65478e4c8392f1d2b4e994e48029cdb54bfb0)
BibTeX
@misc{dce73b4fda0414,
title = {Evidence-Bounded Mental Health Reasoning from Heterogeneous Speech Protocols — arXiv:2608.31014},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.31014},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_e73b4fda0414; sha256 e4bb4b791713ad9e1c0d727cd8e65478e4c8392f1d2b4e994e48029cdb54bfb0},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.31014. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Computational mental health screening using multimodal speech and text has shown great promise. However, existing models often assume all clinical speech protocols carry equivalent evidentiary validity. In reality, heterogeneous protocols, from free interviews to fixed reading tasks, support fundamentally different evidence. Forcing uniform reasoning flattens these boundaries, causing models to hallucinate symptoms from irrelevant text or overclaim support. Even advanced long chain-of-thought LLMs fail to resolve this issue, as free-form reasoning can exacerbate boundary violations. To address this, we reformulate multimodal screening as an evidence-bounded reasoning problem. We introduce the Evidence Package Benchmark, integrating 1,870 packages across six heterogeneous sources with explicit modality masks and evidence permissions. We further propose EviBound, a protocol-aware evidence control framework. Unlike direct LLM prompting, EviBound uses a profile-aware planner to restrict reasoning scope, orchestrates evidence tools via five-way acoustic consensus, and enforces a boundary critic to suppress unsupported claims. Empirical results show EviBound achieves a held-out test Depression AUROC of 0.8658, exceeding the strongest direct omni-modal baseline by +0.0811 AUROC while maintaining zero claim violations. Our work moves beyond unconstrained accuracy toward evidence-consistent, protocol-aware systems for safer clinical NLP research. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-01 in cs.CL). Categorie arXiv: Computation and Language (cs.CL). Pagina del lavoro: https://arxiv.org/abs/2608.31014 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-17T07:47:52.875Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-01?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).