↓跳过正文

ICLR 2026

MCPMark: A Benchmark for Stress-Testing Realistic and Comprehensive MCP Use ↗ ↖
Zijian Wu
, 
Xiangyan Liu
, 
Xinyuan Zhang
, 
Lingjun Chen
, 
Fanqing Meng
, 
Lingxiao Du
, 
Yiran Zhao
, 
Fanshi Zhang
, 
Yaoqi Ye
, 
Jiawei Wang
, 
Zirui Wang
, 
Jinjie Ni
, 
Yufan Yang
, 
Arvin Xu
, 
Michael Qizhe Shieh
Arxiv Github ICLR 2026
The MCP standardizes how LLMs interact with external systems, forming the foundation for general …