Reuters
Benchmarks
Task NameDataset NameSOTA ResultTrendResults
Reuters ChatGLM 1.0 (test)
97.33TPR@FPR=1%
12
Reuters ChatGPT-turbo 1.0 (test)
0.561TPR@FPR=1%
12
Reuters ChatGPT 1.0 (test)
89.82TPR@FPR=1%
12
Reuters GPT4All 1.0 (test)
71.33TPR @ FPR=1%
12
Reuters ChatGPT
90.27TPR @ FPR=1%
12
Reuters ChatGPT-turbo (test)
98.31TPR@FPR=1%
12
Reuters ChatGLM (test)
99.87TPR@FPR=1%
12
Reuters Dolly (test)
79.07TPR@FPR=1%
12
Reuters (test)
0.857Precision
12
reuters
0.473Ranking Loss
11
reuters
62.27Macro-F1
11
reuters
2.535CV
11
reuters (test)
12.33HL
11
reuters
40.21OE
11
reuters
33.93AP
11
N3-Reuters-128 (out-of-domain)
58.1InKB micro F1
11
Reuters 36 classes (test)
70bACC
11
Reuters-21578
95.06Precision @1
11
REUTERS (full)
75.63Accuracy
11
Reuters
99.1Accuracy
10
Reuters-21578 R (test)
90.69Accuracy
10
Reuters Out-of-Sample
85.27Accuracy
9
Reuters In-Sample
89.4Accuracy
9
Reuters-21578 (Tail)
0.489Micro-F1
9
Reuters-21578 (Medium)
80.25Micro-F1
9