MultiWOZ
Benchmarks
TaskDatasetSOTA resultTrendResults
MultiWOZ 2.1 (test)
60.76Joint Goal Accuracy
105
MultiWOZ 2.2 (test)
66.25Joint Goal Accuracy
98
MultiWOZ 2.1 (test)
79.2Joint Goal Accuracy
88
MultiWOZ (test)
94Task Success Rate
78
MultiWOZ 2.1 (test)
22.19BLEU Score
57
MultiWOZ 2.0 (test)
55.03Joint Goal Accuracy
47
MultiWOZ 2.1
61.29Joint Goal Accuracy
46
MultiWOZ 2.4 (test)
78.2Joint Goal Acc
45
MultiWOZ 2.0 (test)
99.1Inform Rate
37
MultiWOZ 2.0 (test)
57.23Joint Goal Accuracy
29
MultiWOZ (test)
35.1BLEU Score
27
MultiWOZ 2.2 (test)
96.48Inform Rate
23
MultiWOZ 2.0 (test)
97.5Inform Accuracy
22
MultiWOZ 2.0 (test)
95.4Inform Rate
22
MultiWOZ
98.7Win Rate
20
MultiWOZ (test)
32.4Joint Goal Acc
17
MultiWOZ
42State Precision
15
MultiWOZ 2.4 (test)
43.8JGA
15
MultiWOZ Challenge
80.1Profile Success Rate (Prof.)
14
MultiWOZ
85.3Success Rate
13
MultiWOZ 2.0
91.8Inform Rate
13
MultiWOZ 2.4
99Inform
12
MultiWOZ zero-shot 2.1
38.86Attraction Accuracy
11
MultiWOZ 2.0
94.3Inform Score
11
Multiwoz
0.77TSE Score
11