Document Question Answering
Transformers
Safetensors
Vietnamese
internvl_chat
image-feature-extraction
custom_code
Instructions to use YuukiAsuna/Vintern-1B-v2-ViTable-docvqa with libraries, inference providers, notebooks, and local apps. Follow these links to get started.
- Libraries
- Transformers
How to use YuukiAsuna/Vintern-1B-v2-ViTable-docvqa with Transformers:
# Use a pipeline as a high-level helper from transformers import pipeline pipe = pipeline("document-question-answering", model="YuukiAsuna/Vintern-1B-v2-ViTable-docvqa", trust_remote_code=True)# Load model directly from transformers import AutoModel model = AutoModel.from_pretrained("YuukiAsuna/Vintern-1B-v2-ViTable-docvqa", trust_remote_code=True, dtype="auto") - Notebooks
- Google Colab
- Kaggle
Update README.md
Browse files
README.md
CHANGED
|
@@ -24,11 +24,11 @@ Vintern-1B-v2-ViTable-docvqa is a fine-tuned version of the 5CD-AI/Vintern-1B-v2
|
|
| 24 |
|
| 25 |
<div align="center">
|
| 26 |
|
| 27 |
-
| Model
|
| 28 |
-
|-----------------------------|------------------------|------------------------|------------------------|
|
| 29 |
-
| Gemini 1.5 Flash
|
| 30 |
-
| Vintern-1B-v2
|
| 31 |
-
| Vintern-1B-v2-ViTable-
|
| 32 |
|
| 33 |
</div>
|
| 34 |
|
|
|
|
| 24 |
|
| 25 |
<div align="center">
|
| 26 |
|
| 27 |
+
| Model | ANLS | Semantic Similarity | MLLM-as-judge (Gemini) |
|
| 28 |
+
|------------------------------|------------------------|------------------------|------------------------|
|
| 29 |
+
| Gemini 1.5 Flash | 0.35 | 0.56 | 0.40 |
|
| 30 |
+
| Vintern-1B-v2 | 0.04 | 0.45 | 0.50 |
|
| 31 |
+
| Vintern-1B-v2-ViTable-docvqa | **0.50** | **0.71** | **0.59** |
|
| 32 |
|
| 33 |
</div>
|
| 34 |
|