minhy112 commited on
Commit
8322a91
·
verified ·
1 Parent(s): cb3cfe1

Archive verified research artifact: summary.json

Browse files
artifacts/experiments/runs/test-qwen3-mmqa_ti-s17/summary.json ADDED
@@ -0,0 +1,154 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ {
2
+ "checkpoint": null,
3
+ "dataset": "mmqa_ti",
4
+ "method": "qwen3",
5
+ "scores": [
6
+ {
7
+ "complete": true,
8
+ "dataset": "mmqa_ti",
9
+ "evaluator_sha256": "1b95323ffdf36a0e8a8afef28c598e6fd560f6e63aa8d48709e51607e84d6100",
10
+ "groups": {
11
+ "modality:image": {
12
+ "em": 21.518987341772153,
13
+ "f1": 24.151898734177212,
14
+ "n": 79
15
+ },
16
+ "modality:image+text": {
17
+ "em": 8.695652173913043,
18
+ "f1": 8.695652173913043,
19
+ "n": 23
20
+ },
21
+ "modality:text": {
22
+ "em": 29.878048780487806,
23
+ "f1": 36.396341463414636,
24
+ "n": 164
25
+ },
26
+ "type:Compare(Compose(TableQ,ImageQ),Compose(TableQ,TextQ))": {
27
+ "em": 0.0,
28
+ "f1": 0.0,
29
+ "n": 4
30
+ },
31
+ "type:Compose(ImageQ,TextQ)": {
32
+ "em": 14.285714285714286,
33
+ "f1": 14.285714285714286,
34
+ "n": 7
35
+ },
36
+ "type:Compose(TextQ,ImageListQ)": {
37
+ "em": 8.333333333333334,
38
+ "f1": 8.333333333333334,
39
+ "n": 12
40
+ },
41
+ "type:ImageListQ": {
42
+ "em": 0.0,
43
+ "f1": 4.40625,
44
+ "n": 32
45
+ },
46
+ "type:ImageQ": {
47
+ "em": 36.170212765957444,
48
+ "f1": 37.59574468085107,
49
+ "n": 47
50
+ },
51
+ "type:TextQ": {
52
+ "em": 29.878048780487806,
53
+ "f1": 36.396341463414636,
54
+ "n": 164
55
+ }
56
+ },
57
+ "input_sha256": "098c9d84dca21b07da69903ff247c5af18232b5c8a66ff157cef35006e1e9da1",
58
+ "metric_scale": "0-100",
59
+ "n_expected": 266,
60
+ "n_missing": 0,
61
+ "n_predictions": 266,
62
+ "parse_success_count": 266,
63
+ "prediction_sha256": "fe83c246052a3393b8a143143522f8f66c4f6efe493ab7e27f7886f75cedf22a",
64
+ "protocol_sha256": "ca6b0397b6532489a12ad40a3a1d18408c0f52c45a20c15fae188d3a59eae462",
65
+ "scores": {
66
+ "em": 25.56390977443609,
67
+ "f1": 30.36466165413534
68
+ },
69
+ "split": "val_calibration",
70
+ "status_counts": {
71
+ "ok": 266
72
+ },
73
+ "target_sha256": "fe1557c98813401676c470515634238527d397080aee9f96f09c30a47e5fe907"
74
+ },
75
+ {
76
+ "complete": true,
77
+ "dataset": "mmqa_ti",
78
+ "evaluator_sha256": "1b95323ffdf36a0e8a8afef28c598e6fd560f6e63aa8d48709e51607e84d6100",
79
+ "groups": {
80
+ "modality:image": {
81
+ "em": 23.40966921119593,
82
+ "f1": 25.38167938931298,
83
+ "n": 393
84
+ },
85
+ "modality:image+text": {
86
+ "em": 20.338983050847457,
87
+ "f1": 22.228813559322035,
88
+ "n": 118
89
+ },
90
+ "modality:text": {
91
+ "em": 25.73170731707317,
92
+ "f1": 31.079268292682926,
93
+ "n": 820
94
+ },
95
+ "type:Compare(Compose(TableQ,ImageQ),Compose(TableQ,TextQ))": {
96
+ "em": 0.0,
97
+ "f1": 0.0,
98
+ "n": 22
99
+ },
100
+ "type:Compose(ImageQ,TextQ)": {
101
+ "em": 54.54545454545455,
102
+ "f1": 54.54545454545455,
103
+ "n": 33
104
+ },
105
+ "type:Compose(TextQ,ImageListQ)": {
106
+ "em": 8.19672131147541,
107
+ "f1": 11.852459016393443,
108
+ "n": 61
109
+ },
110
+ "type:ImageListQ": {
111
+ "em": 2.5,
112
+ "f1": 6.6125,
113
+ "n": 160
114
+ },
115
+ "type:ImageQ": {
116
+ "em": 37.76824034334764,
117
+ "f1": 38.27038626609442,
118
+ "n": 233
119
+ },
120
+ "type:Intersect(ImageListQ,TextQ)": {
121
+ "em": 50.0,
122
+ "f1": 50.0,
123
+ "n": 2
124
+ },
125
+ "type:TextQ": {
126
+ "em": 25.73170731707317,
127
+ "f1": 31.079268292682926,
128
+ "n": 820
129
+ }
130
+ },
131
+ "input_sha256": "bb92409d451ad352f22a819754ff924c2c92bbfec79df241aa382897095badad",
132
+ "metric_scale": "0-100",
133
+ "n_expected": 1331,
134
+ "n_missing": 0,
135
+ "n_predictions": 1331,
136
+ "parse_success_count": 1322,
137
+ "prediction_sha256": "b017a0688449783f446dd7b8334770af4bdb52b6b5adec65f3e29d22d57ccad5",
138
+ "protocol_sha256": "ca6b0397b6532489a12ad40a3a1d18408c0f52c45a20c15fae188d3a59eae462",
139
+ "scores": {
140
+ "em": 24.567993989481593,
141
+ "f1": 28.612321562734785
142
+ },
143
+ "split": "test",
144
+ "status_counts": {
145
+ "ok": 1331
146
+ },
147
+ "target_sha256": "f9ac83d22210f0300be901ec863f8233e4438dcd99660a7ae628d7d9a290c5a8"
148
+ }
149
+ ],
150
+ "seed": 17,
151
+ "test_feedback_for_training": false,
152
+ "test_used": true,
153
+ "validation_run": "experiments/runs/qwen3-mmqa_ti-s17"
154
+ }