| open jev / mailroom control v1 |
Choice |
100.0 |
100.0 |
82.9 |
|
来源 ↗ |
| dbpedia |
Choice |
97.8 |
98.9 |
83.5 |
|
来源 ↗ |
| s1mb generalization contextual choice |
Choice |
97.4 |
100.0 |
82.6 |
|
来源 ↗ |
| arc |
Choice |
94.1 |
98.5 |
64.4 |
|
来源 ↗ |
| open jev / sponsor segment control v1 |
Choice |
92.3 |
100.0 |
52.2 |
|
来源 ↗ |
| s1mb generalization diverse choice |
Choice |
90.8 |
100.0 |
69.5 |
|
来源 ↗ |
| bbq |
Choice |
90.0 |
98.3 |
53.8 |
|
来源 ↗ |
| s1mb generalization contextual noul |
Noul |
90.0 |
98.0 |
62.0 |
|
来源 ↗ |
| s1mb generalization diverse noul |
Noul |
90.0 |
100.0 |
59.2 |
|
来源 ↗ |
| openbookqa |
Choice |
88.9 |
97.2 |
53.7 |
|
来源 ↗ |
| open jev / browser control v1 |
Choice |
87.0 |
100.0 |
67.4 |
|
来源 ↗ |
| laya / enron spam |
Noul |
86.0 |
100.0 |
54.6 |
|
来源 ↗ |
| miqa |
Choice |
84.1 |
95.5 |
61.9 |
|
来源 ↗ |
| snli |
Choice |
83.9 |
98.4 |
46.0 |
|
来源 ↗ |
| open jev / citation control v1 |
Choice |
83.3 |
100.0 |
40.9 |
|
来源 ↗ |
| open jev / customer control v1 |
Noul |
82.3 |
100.0 |
58.0 |
|
来源 ↗ |
| sdoh nli |
Noul |
82.0 |
92.0 |
60.8 |
|
来源 ↗ |
| open jev / phone extraction control v1 |
Noul |
80.9 |
100.0 |
50.4 |
|
来源 ↗ |
| open jev / ir control v1 |
Noul |
78.6 |
100.0 |
47.3 |
|
来源 ↗ |
| open jev / phone extraction control v1 |
Choice |
78.3 |
100.0 |
32.4 |
|
来源 ↗ |
| laya / ag news |
Choice |
77.6 |
94.0 |
75.4 |
|
来源 ↗ |
| quartz |
Choice |
76.2 |
88.1 |
49.1 |
|
来源 ↗ |
| creak |
Noul |
76.0 |
88.0 |
38.8 |
|
来源 ↗ |
| hans |
Noul |
76.0 |
100.0 |
43.3 |
|
来源 ↗ |
| scitail |
Noul |
76.0 |
98.0 |
53.0 |
|
来源 ↗ |
| open jev / mailroom control v1 |
Noul |
75.0 |
100.0 |
62.5 |
|
来源 ↗ |
| lonli |
Choice |
72.1 |
86.9 |
39.2 |
|
来源 ↗ |
| canttalk |
Noul |
72.0 |
86.0 |
43.9 |
|
来源 ↗ |
| open jev / entity alignment control v1 |
Noul |
70.1 |
100.0 |
35.9 |
|
来源 ↗ |
| ethos |
Noul |
70.0 |
76.0 |
43.7 |
|
来源 ↗ |
| paws |
Noul |
70.0 |
88.0 |
45.7 |
|
来源 ↗ |
| open jev / workflow controls v1 / invoice processing |
Noul |
69.7 |
100.0 |
34.6 |
|
来源 ↗ |
| open jev / vizdoom basic v1 |
Noul |
69.5 |
100.0 |
39.0 |
|
来源 ↗ |
| s1mb generalization diverse score |
Score |
68.2 |
95.3 |
42.8 |
|
来源 ↗ |
| open jev / ir control v1 |
Choice |
68.0 |
88.0 |
44.8 |
|
来源 ↗ |
| open jev / workflow controls v1 / customer service |
Noul |
67.7 |
100.0 |
41.6 |
|
来源 ↗ |
| ethics |
Choice |
66.7 |
83.3 |
20.9 |
|
来源 ↗ |
| hatecheck |
Noul |
66.0 |
98.0 |
47.5 |
|
来源 ↗ |
| qasper |
Noul |
66.0 |
88.0 |
41.6 |
|
来源 ↗ |
| scone |
Choice |
66.0 |
92.0 |
34.2 |
|
来源 ↗ |
| scicite |
Choice |
64.5 |
71.0 |
44.0 |
|
来源 ↗ |
| open jev / workflow controls v1 / security incidents |
Noul |
63.1 |
100.0 |
37.8 |
|
来源 ↗ |
| few nerd |
Choice |
62.7 |
86.4 |
43.9 |
|
来源 ↗ |
| followir robust04 |
Noul |
62.0 |
90.9 |
46.9 |
|
来源 ↗ |
| open jev / silent failure control v1 |
Noul |
62.0 |
100.0 |
38.8 |
|
来源 ↗ |
| open jev / customer control v1 |
Choice |
60.9 |
82.6 |
64.3 |
|
来源 ↗ |
| open jev / amount extraction control v1 |
Choice |
60.8 |
100.0 |
55.0 |
|
来源 ↗ |
| contract nli |
Choice |
59.6 |
76.0 |
31.8 |
|
来源 ↗ |
| s1mb generalization contextual score |
Score |
58.6 |
92.7 |
42.1 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonq |
Noul |
56.9 |
67.2 |
39.5 |
|
来源 ↗ |
| massive |
Choice |
56.1 |
96.7 |
64.5 |
|
来源 ↗ |
| defeasible nli |
Choice |
54.4 |
80.4 |
32.8 |
|
来源 ↗ |
| crows pairs |
Choice |
52.1 |
70.8 |
30.4 |
|
来源 ↗ |
| gretel pii |
Choice |
51.3 |
94.5 |
43.1 |
|
来源 ↗ |
| argument quality |
Choice |
51.0 |
95.9 |
37.4 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanomsmarco |
Noul |
50.9 |
62.5 |
37.3 |
|
来源 ↗ |
| go emotions |
Noul |
48.8 |
61.3 |
38.3 |
|
来源 ↗ |
| synthetic relevance / nanocoir |
Noul |
48.2 |
73.0 |
33.6 |
|
来源 ↗ |
| babi nli |
Noul |
48.0 |
84.0 |
37.0 |
|
来源 ↗ |
| impli |
Noul |
48.0 |
88.0 |
46.7 |
|
来源 ↗ |
| lexcomp |
Noul |
48.0 |
52.0 |
22.5 |
|
来源 ↗ |
| aegis2 |
Noul |
47.9 |
62.3 |
32.8 |
|
来源 ↗ |
| mtop |
Choice |
47.5 |
86.9 |
58.6 |
|
来源 ↗ |
| open jev / amount extraction control v1 |
Noul |
46.9 |
100.0 |
33.3 |
|
来源 ↗ |
| arct |
Choice |
44.7 |
89.4 |
39.4 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanodbpedia |
Noul |
43.7 |
52.1 |
29.0 |
|
来源 ↗ |
| open jev / email selection control v1 |
Choice |
42.6 |
100.0 |
23.4 |
|
来源 ↗ |
| open jev / workflow controls v1 / agent trace observability |
Noul |
42.5 |
100.0 |
27.8 |
|
来源 ↗ |
| open jev / email selection control v1 |
Noul |
42.3 |
100.0 |
40.6 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Choice |
42.1 |
100.0 |
46.2 |
|
来源 ↗ |
| open jev / customer control v1 |
Score |
41.8 |
99.3 |
36.7 |
|
来源 ↗ |
| laya / typed decisions |
Noul |
41.3 |
47.8 |
24.4 |
|
来源 ↗ |
| ethics |
Noul |
41.2 |
68.6 |
23.9 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanofiqa2018 |
Noul |
40.5 |
54.0 |
28.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanotouche2020 |
Score |
40.2 |
44.3 |
20.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanomsmarco |
Score |
40.0 |
52.6 |
17.8 |
|
来源 ↗ |
| laya / typed decisions |
Score |
38.4 |
63.5 |
26.5 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanotouche2020 |
Noul |
38.3 |
47.0 |
26.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanodbpedia |
Score |
37.6 |
44.9 |
17.6 |
|
来源 ↗ |
| temporal nli |
Choice |
37.1 |
80.7 |
18.3 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Score |
35.7 |
90.7 |
27.7 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonfcorpus |
Noul |
35.4 |
46.2 |
25.1 |
|
来源 ↗ |
| open jev / vizdoom basic v1 |
Score |
35.1 |
100.0 |
29.6 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Choice |
35.0 |
95.0 |
28.4 |
|
来源 ↗ |
| winobias |
Choice |
34.8 |
91.3 |
30.1 |
|
来源 ↗ |
| sgd |
Choice |
33.9 |
96.6 |
24.4 |
|
来源 ↗ |
| followir core17 |
Noul |
33.6 |
46.6 |
21.2 |
|
来源 ↗ |
| wiqa |
Choice |
33.3 |
58.7 |
19.3 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanofiqa2018 |
Score |
32.4 |
46.7 |
15.2 |
|
来源 ↗ |
| patent similarity |
Score |
32.3 |
48.3 |
17.7 |
|
来源 ↗ |
| civil comments |
Noul |
31.1 |
36.7 |
21.4 |
|
来源 ↗ |
| ruletaker |
Noul |
30.0 |
96.0 |
27.7 |
|
来源 ↗ |
| hwu64 |
Choice |
29.8 |
92.5 |
56.8 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Noul |
29.5 |
100.0 |
27.2 |
|
来源 ↗ |
| open jev / drone control v1 |
Noul |
29.4 |
100.0 |
50.2 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonfcorpus |
Score |
26.4 |
38.6 |
10.7 |
|
来源 ↗ |
| winowhy |
Noul |
26.0 |
62.0 |
18.8 |
|
来源 ↗ |
| open jev / entity alignment control v1 |
Score |
24.8 |
100.0 |
8.1 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanohotpotqa |
Noul |
23.9 |
70.8 |
28.4 |
|
来源 ↗ |
| ud ewt |
Choice |
23.8 |
56.7 |
13.2 |
|
来源 ↗ |
| fol nli |
Choice |
22.9 |
54.1 |
10.4 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanonq |
Score |
22.3 |
49.4 |
9.3 |
|
来源 ↗ |
| plane |
Noul |
22.0 |
100.0 |
11.1 |
|
来源 ↗ |
| stepgame |
Choice |
21.4 |
82.1 |
17.8 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanohotpotqa |
Score |
21.3 |
63.6 |
13.9 |
|
来源 ↗ |
| open jev / reasoning control v1 |
Noul |
21.0 |
100.0 |
25.3 |
|
来源 ↗ |
| clinc |
Choice |
20.0 |
99.3 |
57.3 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoarguana |
Score |
19.6 |
36.0 |
7.7 |
|
来源 ↗ |
| laya / typed decisions |
Choice |
18.2 |
34.5 |
18.9 |
|
来源 ↗ |
| cladder |
Noul |
18.0 |
94.0 |
17.7 |
|
来源 ↗ |
| logical entailment |
Noul |
18.0 |
72.0 |
12.2 |
|
来源 ↗ |
| banking77 |
Choice |
17.9 |
93.7 |
54.6 |
|
来源 ↗ |
| open jev / drone control v1 |
Choice |
16.7 |
100.0 |
4.8 |
|
来源 ↗ |
| gsm8k |
Choice |
15.7 |
78.6 |
23.5 |
|
来源 ↗ |
| boardgameqa |
Choice |
15.6 |
62.5 |
15.6 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoscidocs |
Noul |
15.6 |
41.7 |
17.3 |
|
来源 ↗ |
| esci |
Choice |
15.1 |
46.6 |
13.0 |
|
来源 ↗ |
| argument quality |
Noul |
13.4 |
23.9 |
5.4 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoscidocs |
Score |
11.9 |
36.7 |
7.2 |
|
来源 ↗ |
| robust lr |
Choice |
11.7 |
66.7 |
18.8 |
|
来源 ↗ |
| spartqa |
Choice |
11.3 |
38.7 |
4.4 |
|
来源 ↗ |
| open jev / snake v1 |
Noul |
10.2 |
100.0 |
9.0 |
|
来源 ↗ |
| synthetic relevance / nanobeir / nanoarguana |
Noul |
9.0 |
48.2 |
15.6 |
|
来源 ↗ |
| synthetic relevance / nanocoir |
Score |
8.4 |
44.3 |
4.3 |
|
来源 ↗ |
| tracie |
Noul |
8.0 |
34.0 |
8.7 |
|
来源 ↗ |
| open jev / ir control v1 |
Score |
7.2 |
98.6 |
10.1 |
|
来源 ↗ |
| nlsat |
Noul |
6.0 |
22.0 |
2.1 |
|
来源 ↗ |
| aqua rat |
Choice |
4.2 |
77.5 |
10.7 |
|
来源 ↗ |
| followir news21 |
Noul |
0.7 |
42.7 |
14.9 |
|
来源 ↗ |
| corr2cause |
Choice |
0.0 |
91.7 |
3.8 |
|
来源 ↗ |
| hh rlhf |
Choice |
0.0 |
23.1 |
3.1 |
|
来源 ↗ |
| open jev / context retention control v1 |
Noul |
0.0 |
100.0 |
19.0 |
|
来源 ↗ |
| open jev / drone control v1 |
Score |
0.0 |
99.2 |
1.1 |
|
来源 ↗ |
| open jev / painting geometry v1 |
Score |
0.0 |
100.0 |
15.2 |
|
来源 ↗ |
| open jev / snake v1 |
Choice |
0.0 |
83.5 |
8.1 |
|
来源 ↗ |
| open jev / tic tac toe v1 |
Choice |
0.0 |
48.5 |
3.5 |
|
来源 ↗ |
| poem sentiment |
Choice |
0.0 |
57.1 |
4.0 |
|
来源 ↗ |