Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Target-guided proactive dialogue generation on DuRecDial ID (test)
Loading...
3.31
Perplexity (PPL)
TPDial
3.2868
3.4434
3.6
3.7566
May 12, 2026
Perplexity (PPL)
W. F1
BLEU-1
BLEU-2
DIST-1
DIST-2
K. F1
Failure Rate
Updated 21d ago
Evaluation Results
Method
Method
Links
Perplexity (PPL)
W. F1
BLEU-1
BLEU-2
DIST-1
DIST-2
K. F1
Failure Rate
TPDial
repro=⋄
2026.05
3.31
40.77
38.7
31.2
1.2
6.6
54.08
23.19
T5-Zh
2026.05
3.37
44.35
41
33
1.1
6.4
49.35
22.3
TRIPDial
repro=⋄
2026.05
3.46
44.24
40.4
32.7
1.2
6.6
54.64
22.6
Our
mode=soft
2026.05
3.87
44.87
43.3
34.8
1
6.1
49.83
17.13
Our
mode=hard
2026.05
3.89
44.7
43.3
34.7
1
6.1
49.62
17.13
Feedback
Search any
task
Search any
task