Popular repositories Loading

1. -SFT-RLHF- -SFT-RLHF- Public

   Python

2. chinese-tool-calling-sft-dpo chinese-tool-calling-sft-dpo Public

   Reproducible Qwen2.5 QLoRA pipeline for Chinese tool-calling fine-tuning with SFT, DPO, xLAM data conversion, Kaggle training, and schema-aware evaluation.

   Python

3. multi-agent-public-opinion-monitor multi-agent-public-opinion-monitor Public

   企业级多 Agent 舆情监测系统模板，支持定时推送、行业情报采集、结构化研判与 Word/DOCX 自动化交付。

   Python

4. langgraph-underwriting-agent langgraph-underwriting-agent Public

   基于 LangGraph 的核保规则开发智能体，支持工具调用、流式响应和会话持久化，并集成规则编辑器。

   Python