| open jev / phone extraction control v1 |
Noul |
100.0 |
100.0 |
56.8 |
|
来源 ↗ |
| dbpedia |
Choice |
98.9 |
98.9 |
85.8 |
|
来源 ↗ |
| s1mb generalization contextual choice |
Choice |
97.4 |
100.0 |
85.8 |
|
来源 ↗ |
| arc |
Choice |
97.1 |
98.5 |
69.3 |
|
来源 ↗ |
| snli |
Choice |
95.2 |
98.4 |
52.9 |
|
来源 ↗ |
| clinc |
Choice |
93.8 |
99.3 |
65.4 |
|
来源 ↗ |
| bbq |
Choice |
91.7 |
98.3 |
60.0 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Choice |
89.5 |
100.0 |
52.6 |
|
来源 ↗ |
| miqa |
Choice |
88.6 |
95.5 |
66.9 |
|
来源 ↗ |
| open jev / mailroom control v1 |
Noul |
88.1 |
100.0 |
68.0 |
|
来源 ↗ |
| open jev / browser control v1 |
Choice |
87.0 |
100.0 |
72.2 |
|
来源 ↗ |
| laya / ag news |
Choice |
86.6 |
94.0 |
75.3 |
|
来源 ↗ |
| openbookqa |
Choice |
86.1 |
97.2 |
58.7 |
|
来源 ↗ |
| s1mb generalization diverse choice |
Choice |
84.1 |
100.0 |
74.0 |
|
来源 ↗ |
| sdoh nli |
Noul |
84.0 |
94.0 |
64.7 |
|
来源 ↗ |
| ethics |
Choice |
83.3 |
83.3 |
24.0 |
|
来源 ↗ |
| open jev / mailroom control v1 |
Choice |
83.3 |
100.0 |
86.3 |
|
来源 ↗ |
| massive |
Choice |
82.5 |
96.7 |
69.3 |
|
来源 ↗ |
| open jev / drone control v1 |
Noul |
82.3 |
100.0 |
58.5 |
|
来源 ↗ |
| hatecheck |
Noul |
82.0 |
98.0 |
51.7 |
|
来源 ↗ |
| s1mb generalization contextual noul |
Noul |
82.0 |
98.0 |
67.8 |
|
来源 ↗ |
| s1mb generalization diverse noul |
Noul |
82.0 |
100.0 |
65.8 |
|
来源 ↗ |
| gretel pii |
Choice |
81.2 |
94.5 |
47.8 |
|
来源 ↗ |
| mtop |
Choice |
81.2 |
88.6 |
63.3 |
|
来源 ↗ |
| banking77 |
Choice |
80.0 |
93.7 |
60.0 |
|
来源 ↗ |
| followir robust04 |
Noul |
78.7 |
90.9 |
51.1 |
|
来源 ↗ |
| open jev / customer control v1 |
Choice |
78.3 |
82.6 |
68.4 |
|
来源 ↗ |
| scitail |
Noul |
78.0 |
98.0 |
57.0 |
|
来源 ↗ |
| open jev / sponsor segment control v1 |
Choice |
75.6 |
100.0 |
60.1 |
|
来源 ↗ |
| ethos |
Noul |
74.0 |
76.0 |
46.8 |
|
来源 ↗ |
| impli |
Noul |
74.0 |
88.0 |
51.5 |
|
来源 ↗ |
| quartz |
Choice |
73.8 |
90.5 |
54.4 |
|
来源 ↗ |
| lonli |
Choice |
73.8 |
96.7 |
45.7 |
|
来源 ↗ |
| hwu64 |
Choice |
73.4 |
93.6 |
62.9 |
|
来源 ↗ |
| few nerd |
Choice |
73.1 |
86.4 |
48.5 |
|
来源 ↗ |
| open jev / email selection control v1 |
Noul |
73.1 |
100.0 |
48.5 |
|
来源 ↗ |
| laya / enron spam |
Noul |
72.0 |
100.0 |
60.4 |
|
来源 ↗ |
| arct |
Choice |
70.2 |
89.4 |
43.2 |
|
来源 ↗ |
| scone |
Choice |
70.0 |
94.0 |
40.7 |
|
来源 ↗ |
| open jev / workflow controls v1 / customer service |
Noul |
68.8 |
100.0 |
48.8 |
|
来源 ↗ |
| qasper |
Noul |
66.0 |
88.0 |
46.9 |
|
来源 ↗ |
| open jev / amount extraction control v1 |
Choice |
64.7 |
100.0 |
58.6 |
|
来源 ↗ |
| open jev / ir control v1 |
Noul |
64.3 |
100.0 |
54.2 |
|
来源 ↗ |
| s1mb generalization contextual score |
Score |
64.2 |
97.3 |
46.7 |
|
来源 ↗ |
| hans |
Noul |
64.0 |
100.0 |
47.1 |
|
来源 ↗ |
| argument quality |
Choice |
63.3 |
95.9 |
45.0 |
|
来源 ↗ |
| open jev / workflow controls v1 / invoice processing |
Noul |
61.2 |
100.0 |
40.7 |
|
来源 ↗ |
| s1mb generalization diverse score |
Score |
60.3 |
96.3 |
46.8 |
|
来源 ↗ |
| scicite |
Choice |
59.7 |
71.0 |
45.8 |
|
来源 ↗ |
| paws |
Noul |
58.0 |
90.0 |
48.5 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonq |
Noul |
57.9 |
71.2 |
44.1 |
|
来源 ↗ |
| crows pairs |
Choice |
56.3 |
70.8 |
33.6 |
|
来源 ↗ |
| canttalk |
Noul |
56.0 |
92.0 |
47.9 |
|
来源 ↗ |
| creak |
Noul |
56.0 |
90.0 |
42.8 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanomsmarco |
Noul |
54.6 |
63.0 |
41.5 |
|
来源 ↗ |
| babi nli |
Noul |
54.0 |
92.0 |
42.1 |
|
来源 ↗ |
| winobias |
Choice |
52.2 |
93.5 |
35.7 |
|
来源 ↗ |
| contract nli |
Choice |
50.9 |
80.3 |
37.3 |
|
来源 ↗ |
| open jev / vizdoom basic v1 |
Noul |
50.7 |
100.0 |
45.8 |
|
来源 ↗ |
| defeasible nli |
Choice |
50.0 |
87.0 |
37.3 |
|
来源 ↗ |
| aegis2 |
Noul |
49.5 |
64.9 |
36.3 |
|
来源 ↗ |
| go emotions |
Noul |
49.0 |
63.4 |
40.4 |
|
来源 ↗ |
| synthetic relevance / nanocoir |
Noul |
46.3 |
73.0 |
38.3 |
|
来源 ↗ |
| open jev / customer control v1 |
Score |
46.1 |
99.3 |
39.9 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanofiqa2018 |
Noul |
45.0 |
58.0 |
32.0 |
|
来源 ↗ |
| open jev / ir control v1 |
Choice |
44.0 |
88.0 |
50.8 |
|
来源 ↗ |
| open jev / workflow controls v1 / security incidents |
Noul |
44.0 |
100.0 |
43.2 |
|
来源 ↗ |
| open jev / entity alignment control v1 |
Noul |
42.3 |
100.0 |
43.7 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanodbpedia |
Noul |
41.7 |
52.1 |
32.1 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Noul |
41.0 |
100.0 |
34.6 |
|
来源 ↗ |
| lexcomp |
Noul |
40.0 |
60.0 |
24.5 |
|
来源 ↗ |
| open jev / vizdoom basic v1 |
Score |
39.6 |
100.0 |
35.1 |
|
来源 ↗ |
| open jev / phone extraction control v1 |
Choice |
39.1 |
100.0 |
38.6 |
|
来源 ↗ |
| open jev / workflow controls v1 / agent trace observability |
Noul |
38.1 |
100.0 |
33.3 |
|
来源 ↗ |
| ruletaker |
Noul |
38.0 |
100.0 |
31.9 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanomsmarco |
Score |
37.7 |
54.9 |
20.9 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanodbpedia |
Score |
35.9 |
45.0 |
20.1 |
|
来源 ↗ |
| laya / typed decisions |
Score |
35.7 |
76.0 |
31.9 |
|
来源 ↗ |
| followir core17 |
Noul |
35.3 |
46.6 |
23.5 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Choice |
35.0 |
100.0 |
34.7 |
|
来源 ↗ |
| ethics |
Noul |
32.1 |
75.6 |
28.0 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanohotpotqa |
Noul |
31.3 |
77.7 |
31.7 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonfcorpus |
Noul |
31.2 |
49.7 |
28.0 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanotouche2020 |
Noul |
31.0 |
47.0 |
29.3 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanotouche2020 |
Score |
31.0 |
47.5 |
22.9 |
|
来源 ↗ |
| patent similarity |
Score |
30.6 |
48.3 |
20.3 |
|
来源 ↗ |
| open jev / citation control v1 |
Choice |
30.3 |
100.0 |
48.5 |
|
来源 ↗ |
| cladder |
Noul |
30.0 |
94.0 |
22.1 |
|
来源 ↗ |
| tracie |
Noul |
30.0 |
58.0 |
10.0 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanofiqa2018 |
Score |
29.9 |
46.7 |
17.4 |
|
来源 ↗ |
| sgd |
Choice |
28.8 |
96.6 |
26.1 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Noul |
28.7 |
100.0 |
28.9 |
|
来源 ↗ |
| stepgame |
Choice |
28.6 |
82.1 |
20.5 |
|
来源 ↗ |
| open jev / amount extraction control v1 |
Noul |
28.1 |
100.0 |
37.9 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Score |
27.5 |
95.1 |
31.7 |
|
来源 ↗ |
| esci |
Choice |
27.4 |
46.6 |
14.8 |
|
来源 ↗ |
| ud ewt |
Choice |
25.8 |
64.1 |
15.3 |
|
来源 ↗ |
| gsm8k |
Choice |
25.7 |
82.9 |
30.0 |
|
来源 ↗ |
| laya / typed decisions |
Choice |
25.7 |
34.5 |
21.4 |
|
来源 ↗ |
| temporal nli |
Choice |
24.2 |
83.9 |
20.9 |
|
来源 ↗ |
| robust lr |
Choice |
23.3 |
66.7 |
22.8 |
|
来源 ↗ |
| wiqa |
Choice |
22.2 |
60.3 |
21.0 |
|
来源 ↗ |
| laya / typed decisions |
Noul |
20.5 |
49.2 |
27.3 |
|
来源 ↗ |
| boardgameqa |
Choice |
20.3 |
75.0 |
19.2 |
|
来源 ↗ |
| civil comments |
Noul |
18.3 |
39.8 |
23.0 |
|
来源 ↗ |
| open jev / context retention control v1 |
Noul |
18.0 |
100.0 |
26.7 |
|
来源 ↗ |
| fol nli |
Choice |
16.4 |
68.8 |
12.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonfcorpus |
Score |
15.5 |
38.6 |
12.8 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoscidocs |
Noul |
15.1 |
46.6 |
20.2 |
|
来源 ↗ |
| nlsat |
Noul |
14.0 |
22.0 |
2.8 |
|
来源 ↗ |
| winowhy |
Noul |
12.0 |
64.0 |
21.3 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanohotpotqa |
Score |
12.0 |
63.6 |
16.0 |
|
来源 ↗ |
| open jev / snake v1 |
Noul |
11.7 |
100.0 |
13.2 |
|
来源 ↗ |
| aqua rat |
Choice |
11.3 |
77.5 |
14.0 |
|
来源 ↗ |
| followir news21 |
Noul |
10.7 |
48.0 |
17.3 |
|
来源 ↗ |
| spartqa |
Choice |
9.7 |
91.9 |
6.8 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoscidocs |
Score |
9.5 |
36.7 |
8.3 |
|
来源 ↗ |
| logical entailment |
Noul |
8.0 |
72.0 |
13.4 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonq |
Score |
7.8 |
49.4 |
10.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoarguana |
Noul |
6.9 |
53.5 |
18.8 |
|
来源 ↗ |
| open jev / snake v1 |
Choice |
6.2 |
83.5 |
12.6 |
|
来源 ↗ |
| argument quality |
Noul |
3.7 |
26.9 |
6.4 |
|
来源 ↗ |
| corr2cause |
Choice |
3.3 |
91.7 |
4.3 |
|
来源 ↗ |
| plane |
Noul |
2.0 |
100.0 |
11.3 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Score |
0.1 |
100.0 |
21.3 |
|
来源 ↗ |
| hh rlhf |
Choice |
0.0 |
33.3 |
3.3 |
|
来源 ↗ |
| open jev / customer control v1 |
Noul |
0.0 |
100.0 |
62.2 |
|
来源 ↗ |
| open jev / drone control v1 |
Choice |
0.0 |
100.0 |
9.2 |
|
来源 ↗ |
| open jev / drone control v1 |
Score |
0.0 |
99.2 |
3.8 |
|
来源 ↗ |
| open jev / email selection control v1 |
Choice |
0.0 |
100.0 |
30.4 |
|
来源 ↗ |
| open jev / entity alignment control v1 |
Score |
0.0 |
100.0 |
12.1 |
|
来源 ↗ |
| open jev / ir control v1 |
Score |
0.0 |
99.2 |
15.5 |
|
来源 ↗ |
| open jev / silent failure control v1 |
Noul |
0.0 |
100.0 |
43.1 |
|
来源 ↗ |
| open jev / tic tac toe v1 |
Choice |
0.0 |
54.5 |
4.0 |
|
来源 ↗ |
| poem sentiment |
Choice |
0.0 |
68.6 |
5.3 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoarguana |
Score |
0.0 |
43.1 |
8.7 |
|
来源 ↗ |
| synthetic relevance / nanocoir |
Score |
0.0 |
44.3 |
5.0 |
|
来源 ↗ |