Metraipolis › archive › dc_3e7a45d0aeca
Explainable Suicide Risk Assessment on Social Media with Multi-Task QLoRA — arXiv:2610.00610
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.00610
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-06T13:54:04.323Z
- SHA-256 of the archived text
c9cfd6413dcedaaaac206db2cbbf0e6655f055c665ed7d8f7df14c86596a7ab7- Stable URL of this page
https://metraipolis.com/documento/dc_3e7a45d0aeca
How to cite
Explainable Suicide Risk Assessment on Social Media with Multi-Task QLoRA — arXiv:2610.00610 — arXiv (en), https://arxiv.org/abs/2610.00610 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_3e7a45d0aeca (Metraipolis archive dc_3e7a45d0aeca, sha256 c9cfd6413dcedaaaac206db2cbbf0e6655f055c665ed7d8f7df14c86596a7ab7)
BibTeX
@misc{dc3e7a45d0aeca,
title = {Explainable Suicide Risk Assessment on Social Media with Multi-Task QLoRA — arXiv:2610.00610},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.00610},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_3e7a45d0aeca; sha256 c9cfd6413dcedaaaac206db2cbbf0e6655f055c665ed7d8f7df14c86596a7ab7},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.00610. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Explainable suicide-risk assessment requires models not only to estimate risk severity, but also to identify supporting language and the risk and protective factors expressed in a post. We present our system for the IEEE BigData 2026 Cup on Explainable Suicide Risk Assessment on Social Media, which addresses three tasks: risk-level classification, evidence phrase extraction, and multi-label factor identification. Our approach adapts Qwen2.5-Instruct models using quantized low-rank adaptation (QLoRA) and an answer-masked causal language-model objective. We jointly train across all three tasks for risk classification, jointly train on Tasks 1a and 1b for evidence extraction, and adapt Task 2 separately for factor identification. We also tailor aggregation to each output: we average risk-level probabilities from the 32B and 72B models, combine evidence phrases through cross-fold consensus, and calibrate factor-specific decisions through rate matching based on out-of-fold operating points. On the official leaderboard, the final system achieved a composite score of 0.7738, with 0.8089 on Task 1 and 0.6919 on Task 2. Across the evaluated configurations, three-task training performed best for Task 1a, joint training on Tasks 1a and 1b performed best for Task 1b, and task-specific training performed best for Task 2. Probability averaging further improved Task 1a when component models had complementary errors. These findings highlight the value of tailoring both training objectives and aggregation strategies to the output structure of each task within a unified language-model framework. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-10-02 in cs.CL). Categorie arXiv: Computation and Language (cs.CL) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2610.00610 — 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 2026-10-06T12:15:34.626Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-10-02?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).