Introduces MENTAL-TRUST, a seven-level expert-annotated trust dataset for counseling dialogues, and benchmarks 14 models, reporting that fine-tuned smaller models outperform zero-shot LLMs.
Title resolution pending
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
-
Trust Modeling in Counseling Conversations: A Benchmark Study
Introduces MENTAL-TRUST, a seven-level expert-annotated trust dataset for counseling dialogues, and benchmarks 14 models, reporting that fine-tuned smaller models outperform zero-shot LLMs.