A 7B Chinese psychological LLM trained with curated reasoning questions and empathetic dialogues via SFT plus GRPO approaches DeepSeek-R1's exam accuracy at a fraction of the size.
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
-
Psyche-R1: Towards Reliable Psychological LLMs through Unified Empathy, Expertise, and Reasoning
A 7B Chinese psychological LLM trained with curated reasoning questions and empathetic dialogues via SFT plus GRPO approaches DeepSeek-R1's exam accuracy at a fraction of the size.