Hugging Face's logo

ob11
/
Qwen-VL-PRM-7B

Image-to-Text
Transformers
Safetensors
qwen2_5_vl
text-generation-inference
Model card Files
xet
Community
1
nielsr HF Staff commited on
Commit
c8cf0b1
·
verified ·
1 Parent(s): 793b6e1

Fix license typo and add pipeline tag

Browse files

This PR improves the model card for the `Qwen-VL-PRM-7B` model by:

- Correcting the typo in the `licence` metadata tag to `license`.
- Adding the `pipeline_tag: image-text-to-text` to accurately categorize the model for multimodal reasoning tasks, enhancing discoverability on the Hub.

These updates ensure the model card is more accurate and discoverable.

Files changed (1) hide show
  1. README.md +25 -24
README.md CHANGED
@@ -1,10 +1,11 @@
1
  ---
2
  base_model: Qwen/Qwen2.5-VL-7B-Instruct
3
- library_name: transformers
4
- model_name: ob11/Qwen-VL-PRM-7B
5
- licence: apache-2.0
6
  datasets:
7
  - ob11/VL-PRM300K-V1-train
 
 
 
 
8
  ---
9
 
10
  # Model Summary
@@ -21,30 +22,30 @@ The model usage is documented [here](https://github.com/theogbrand/vlprm/blob/ma
21
  # Evaluation
22
  ### Commercial Models
23
  | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
24
- |-------|------|-----------|---------------|-----------|------------|---------|
25
- | GPT-4o | 70.7 | 60.0 | 57.8 | 30.9 | 31.2 | 50.1 |
26
- | o1 | 78.2 | 78.9 | 54.4 | 73.9 | 60.3 | 69.1 |
27
  | o3 | 82.9 | 84.1 | 62.3 | 86.8 | -- | -- |
28
  ### Qwen-2.5-VL Family
29
- | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
30
- |-------|------|-----------|---------------|-----------|------------|---------|
31
- | **Qwen-2.5-VL-3B** | 51.7 | 34.5 | 25.7 | 60.0 | 21.2 | 38.6 |
32
- | + VL-PRM-7B | 53.7 (+2.0) | 44.9 (+10.5) | 28.3 (+2.6) | 64.1 (+4.1) | 21.8 (+0.6) | 42.6 (+4.0) |
33
- | **Qwen-2.5-VL-7B** | 55.0 | 48.0 | 29.1 | 67.8 | 24.2 | 44.8 |
34
- | + VL-PRM-3B | 57.6 (+2.6) | 55.5 (+7.5) | 33.8 (+4.7) | 70.0 (+2.2) | 26.1 (+1.9) | 48.6 (+3.6) |
35
- | + VL-PRM-7B | 57.4 (+2.4) | 54.8 (+6.8) | 35.3 (+6.2) | 71.0 (+3.2) | 26.2 (+2.0) | 48.9 (+4.1) |
36
- | **Qwen-2.5-VL-32B** | 66.0 | 46.2 | 26.9 | 76.9 | 36.7 | 50.5 |
37
- | + VL-PRM-3B | 67.0 (+1.0) | 67.1 (+20.8) | 41.6 (+14.7) | 77.7 (+0.8) | 40.5 (+3.8) | 58.7 (+8.2) |
38
- | + VL-PRM-7B | 67.6 (+1.6) | 66.8 (+20.6) | 44.2 (+17.3) | 78.3 (+1.4) | 40.1 (+3.2) | 59.4 (+8.9) |
39
  ### Gemma-3 Family
40
- | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
41
- |-------|------|-----------|---------------|-----------|------------|---------|
42
- | **Gemma-3-12B** | 57.6 | 45.0 | 29.1 | 58.9 | 28.1 | 43.7 |
43
- | + VL-PRM-3B | 60.4 (+2.8) | 57.7 (+12.7) | 39.7 (+10.6) | 60.3 (+1.4) | 33.8 (+5.7) | 50.4 (+6.7) |
44
- | + VL-PRM-7B | 60.2 (+2.6) | 59.0 (+12.0) | 41.1 (+4.5) | 63.3 (+4.4) | 33.9 (+5.8) | 51.5 (+7.8) |
45
- | **Gemma-3-27B** | 62.9 | 50.8 | 29.9 | 61.6 | 32.4 | 47.5 |
46
- | + VL-PRM-3B | 65.5 (+2.6) | 67.4 (+16.6) | 40.3 (+10.4) | 65.4 (+3.8) | 39.8 (+7.4) | 55.7 (+8.2) |
47
- | + VL-PRM-7B | 64.5 (+1.6) | 67.6 (+16.8) | 41.1 (+11.2) | 65.2 (+3.6) | 40.9 (+8.5) | 55.9 (+8.4) |
48
  ### Framework versions
49
 
50
  - TRL: 0.19.1
 
1
  ---
2
  base_model: Qwen/Qwen2.5-VL-7B-Instruct
 
 
 
3
  datasets:
4
  - ob11/VL-PRM300K-V1-train
5
+ library_name: transformers
6
+ model_name: ob11/Qwen-VL-PRM-7B
7
+ license: apache-2.0
8
+ pipeline_tag: image-text-to-text
9
  ---
10
 
11
  # Model Summary
 
22
  # Evaluation
23
  ### Commercial Models
24
  | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |
25
+ |-------|------|-----------|---------------|-----------|------------|---------|\
26
+ | GPT-4o | 70.7 | 60.0 | 57.8 | 30.9 | 31.2 | 50.1 |\
27
+ | o1 | 78.2 | 78.9 | 54.4 | 73.9 | 60.3 | 69.1 |\
28
  | o3 | 82.9 | 84.1 | 62.3 | 86.8 | -- | -- |
29
  ### Qwen-2.5-VL Family
30
+ | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |\
31
+ |-------|------|-----------|---------------|-----------|------------|---------|\
32
+ | **Qwen-2.5-VL-3B** | 51.7 | 34.5 | 25.7 | 60.0 | 21.2 | 38.6 |\
33
+ | + VL-PRM-7B | 53.7 (+2.0) | 44.9 (+10.5) | 28.3 (+2.6) | 64.1 (+4.1) | 21.8 (+0.6) | 42.6 (+4.0) |\
34
+ | **Qwen-2.5-VL-7B** | 55.0 | 48.0 | 29.1 | 67.8 | 24.2 | 44.8 |\
35
+ | + VL-PRM-3B | 57.6 (+2.6) | 55.5 (+7.5) | 33.8 (+4.7) | 70.0 (+2.2) | 26.1 (+1.9) | 48.6 (+3.6) |\
36
+ | + VL-PRM-7B | 57.4 (+2.4) | 54.8 (+6.8) | 35.3 (+6.2) | 71.0 (+3.2) | 26.2 (+2.0) | 48.9 (+4.1) |\
37
+ | **Qwen-2.5-VL-32B** | 66.0 | 46.2 | 26.9 | 76.9 | 36.7 | 50.5 |\
38
+ | + VL-PRM-3B | 67.0 (+1.0) | 67.1 (+20.8) | 41.6 (+14.7) | 77.7 (+0.8) | 40.5 (+3.8) | 58.7 (+8.2) |\
39
+ | + VL-PRM-7B | 67.6 (+1.6) | 66.8 (+20.6) | 44.2 (+17.3) | 78.3 (+1.4) | 40.1 (+3.2) | 59.4 (+8.9) |\
40
  ### Gemma-3 Family
41
+ | Model | MMMU | PuzzleVQA | AlgoPuzzleVQA | MathVista | MathVision | Overall |\
42
+ |-------|------|-----------|---------------|-----------|------------|---------|\
43
+ | **Gemma-3-12B** | 57.6 | 45.0 | 29.1 | 58.9 | 28.1 | 43.7 |\
44
+ | + VL-PRM-3B | 60.4 (+2.8) | 57.7 (+12.7) | 39.7 (+10.6) | 60.3 (+1.4) | 33.8 (+5.7) | 50.4 (+6.7) |\
45
+ | + VL-PRM-7B | 60.2 (+2.6) | 59.0 (+12.0) | 41.1 (+4.5) | 63.3 (+4.4) | 33.9 (+5.8) | 51.5 (+7.8) |\
46
+ | **Gemma-3-27B** | 62.9 | 50.8 | 29.9 | 61.6 | 32.4 | 47.5 |\
47
+ | + VL-PRM-3B | 65.5 (+2.6) | 67.4 (+16.6) | 40.3 (+10.4) | 65.4 (+3.8) | 39.8 (+7.4) | 55.7 (+8.2) |\
48
+ | + VL-PRM-7B | 64.5 (+1.6) | 67.6 (+16.8) | 41.1 (+11.2) | 65.2 (+3.6) | 40.9 (+8.5) | 55.9 (+8.4) |\
49
  ### Framework versions
50
 
51
  - TRL: 0.19.1