Human Preference Evaluation
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
English-to-Code-Switched human preference evaluation (Out of domain)
434.5Score
6
Human Preference Evaluation 55 prompts
500Votes
6
Human Preference Evaluation
19.27Preference Score (Baseline)
4
Human Preference Evaluation 371 prompts (test)
39.89Recall @1
3
Human Preference Evaluation 466 prompts (test)
65.14Preference Accuracy
3