MT-Bench

Dataset Information
Modalities
Texts
Languages
English
Introduced
2023
License
Unknown

Overview

This dataset contains 3.3K expert-level pairwise human preferences for model responses generated by 6 models in response to 80 MT-bench questions. The 6 models are GPT-4, GPT-3.5, Claud-v1, Vicuna-13B, Alpaca-13B, and LLaMA-13B. The annotators are mostly graduate students with expertise in the topic areas of each of the questions.

Variants: MT-Bench

Associated Benchmarks

This dataset is used in 1 benchmark:

Recent Benchmark Submissions

No recent benchmark submissions available for this dataset.

Research Papers

No papers with results on this dataset found.