search

THUDM / AgentBench

A Comprehensive Benchmark to Evaluate LLMs as Agents (ICLR'24)
https://llmbench.ai
Apache License 2.0
2.23k stars 159 forks source link

fix: fix AgentBench/data/os_interaction/data/4/ N11.json #149

Open minleminzui opened 4 months ago

minleminzui commented 4 months ago

the reference answer doesn't following the description