A unified evaluation framework for large language models

ScienceWorld is a text-based virtual environment centered around accomplishing tasks from the standardized elementary science curriculum.

An extensible benchmark for evaluating large language models on planning

An Analytical Evaluation Board of Multi-turn LLM Agents [NeurIPS 2024 Oral]

Benchmark LLMs by fighting in Street Fighter 3! The new way to evaluate the quality of an LLM