Hugging Face's logo

Vision-CAIR
/
LongVU_Qwen2_7B

Video-Text-to-Text
Safetensors
cambrian_qwen
Eval Results
Model card Files Community
6
Vision-CAIR commited on
Commit
d69cfa6
•
1 Parent(s): 5b15eaf

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +43 -42
README.md CHANGED
@@ -4,49 +4,50 @@ datasets:
4
  - shenxq/VideoChat2
5
  base_model:
6
  - Qwen/Qwen2-7B-Instruct
 
7
  model-index:
8
- - name: llava-onevision-qwen-7b-ov
9
- results:
10
- - task:
11
- type: multimodal
12
- dataset:
13
- name: EgoSchema
14
- type: egoschema
15
- metrics:
16
- - type: accuracy
17
- value: 67.6
18
- name: accuracy
19
- verified: true
20
- - task:
21
- type: multimodal
22
- dataset:
23
- name: MLVU
24
- type: mlvu
25
- metrics:
26
- - type: accuracy
27
- value: 65.4
28
- name: accuracy
29
- verified: true
30
- - task:
31
- type: multimodal
32
- dataset:
33
- name: MVBench
34
- type: mvbench
35
- metrics:
36
- - type: accuracy
37
- value: 66.9
38
- name: accuracy
39
- verified: true
40
- - task:
41
- type: multimodal
42
- dataset:
43
- name: VideoMME
44
- type: videomme
45
- metrics:
46
- - type: accuracy
47
- value: 60.6
48
- name: accuracy
49
- verified: true
50
  ---
51
  # LongVU
52
 
 
4
  - shenxq/VideoChat2
5
  base_model:
6
  - Qwen/Qwen2-7B-Instruct
7
+ - Vision-CAIR/LongVU_Qwen2_7B_img
8
  model-index:
9
+ - name: llava-onevision-qwen-7b-ov
10
+ results:
11
+ - task:
12
+ type: multimodal
13
+ dataset:
14
+ name: EgoSchema
15
+ type: egoschema
16
+ metrics:
17
+ - type: accuracy
18
+ value: 67.6
19
+ name: accuracy
20
+ verified: true
21
+ - task:
22
+ type: multimodal
23
+ dataset:
24
+ name: MLVU
25
+ type: mlvu
26
+ metrics:
27
+ - type: accuracy
28
+ value: 65.4
29
+ name: accuracy
30
+ verified: true
31
+ - task:
32
+ type: multimodal
33
+ dataset:
34
+ name: MVBench
35
+ type: mvbench
36
+ metrics:
37
+ - type: accuracy
38
+ value: 66.9
39
+ name: accuracy
40
+ verified: true
41
+ - task:
42
+ type: multimodal
43
+ dataset:
44
+ name: VideoMME
45
+ type: videomme
46
+ metrics:
47
+ - type: accuracy
48
+ value: 60.6
49
+ name: accuracy
50
+ verified: true
51
  ---
52
  # LongVU
53