AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents
KDD 2026 (CCF-A). A benchmark for diagnosing step-level process quality in tool-using agents.
KDD 2026 (CCF-A). A benchmark for diagnosing step-level process quality in tool-using agents.
KDD 2026 (CCF-A). A benchmark for diagnosing step-level process quality in tool-using agents.
Reflective reinforcement learning for tool learning.
Reflective reinforcement learning for tool learning.
Project Lead — edge-scale (8B) agent LLM mastering MCP tools; surpasses GPT-4o and Qwen3-8B; 5,500+ downloads
ArXiv 2025. Team contribution (led MCP agent capabilities). 10,700+ GitHub Stars.
ArXiv 2025. Team contribution (led MCP agent capabilities). 10,700+ GitHub Stars.