TLDR AI · ENGINEERING & RESOURCES

EVALUATION BOARD OF MULTI-TURN LLM

2024-01-26hkust-nlp.github.io3 min read

Summary from the TLDR AI newsletter

AgentBoard is a benchmark designed for multi-turn LLM agents, complemented by an analytical evaluation board for detailed model assessment beyond final success rates.

Read the original article ↗View the archived TLDR AI issue