Projects项目
Open-source benchmarks, sandboxes, and tools I build and maintain around neuro-symbolic reasoning and LLM agents.
我围绕神经符号推理与大模型智能体构建并维护的开源基准、沙盒与工具。
ChinaTravel
An open-ended benchmark and sandbox for realistic multi-day, multi-POI travel planning with compositional constraint validation for language agents.
一个面向语言智能体的开放式基准与沙盒，用于真实的多日、多景点旅行规划，并支持组合式约束校验。
Neuro-Symbolic AI神经符号 AILLM Agents大模型智能体