Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Knowledge-intensive language tasks evaluation on Knowledge Intensive Tasks (test)

43.9NQ

RA-DIT 65B

3.65214.10124.5534.999Oct 2, 2023
Updated 1mo ago

Evaluation Results

MethodLinks
2023.10
43.964.975.123.240.790.755.872.468.417.351.855.2
2023.10
43.5-72.8-36.686.980.578.172.815.7-60.9
2023.10
42.4-74.5-34.787.166.574.958.915.5-56.8
2023.10
42.364.474.922.841.189.446.460.468.916.851.152.7
2023.10
35.264.675.421.239.780.745.173.753.116.449.150.5
2023.10
31.663.471.822.122.681.548.239.452.117.447.245
2023.10
28.859.772.619.13273.311.850.836.316.145.143.1
2023.10
5.251.255.819.512.559.30.66.71.315.632.922.8