HyperAI
Command Palette
Search for a command to run...
Leistungsergebnisse verschiedener Modelle auf diesem Benchmark
Metriken
# of docs
# of the queries
average perspectives
# of the perspectives
average length of docs
supporting perspectives
undermining perspectives
Cover@10
Cover@50
Cover@100
Recall@10
Recall@50
Recall@100
GPT-4 Score
Cover@20
Recall@20
3 Zeilen insgesamt