LLM-generated English summaries of Chinese speech can train English models to predict suicidal risk about as well as Chinese models trained on Chinese summaries, at least on the development set.
A high-level architecture of our approach is shown in Figure 1
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Language-Agnostic Suicidal Risk Detection Using Large Language Models
LLM-generated English summaries of Chinese speech can train English models to predict suicidal risk about as well as Chinese models trained on Chinese summaries, at least on the development set.