The open-source model layoutlmv2-large-uncased-finetuned-vi-infovqa - Empowering Vietnamese document information extraction and question answering

Layoutlmv2 Large Uncased Finetuned Vi Infovqa

Developed by tiennvcs

A document visual question answering model fine-tuned based on microsoft/layoutlmv2-large-uncased, suitable for Vietnamese information extraction tasks

Text-to-Image

Transformers

#Document Visual Question Answering #Vietnamese Information Extraction #Multimodal Pre-training

Downloads 16

Release Time : 3/2/2022

Model Overview

This model is a LayoutLMv2 model optimized for document visual question answering (VQA) tasks, specifically adapted for Vietnamese information extraction scenarios, capable of understanding document layout and visual information for question answering

Model Features

Multimodal Understanding Capability

Combines text, layout, and visual information for comprehensive understanding

Vietnamese Optimization

Specifically fine-tuned for Vietnamese document information extraction tasks

Document Structure Awareness

Capable of understanding document layout and structural information

Model Capabilities

Document Visual Question Answering

Vietnamese Information Extraction

Document Layout Analysis

Multimodal Understanding

Use Cases

Document Processing

Vietnamese Form Information Extraction

Automatically extracts key information from Vietnamese form documents

Document Visual Question Answering System

Answers natural language questions about document content

Training Loss	Epoch	Step	Validation Loss
No log	0.17	100	4.6181
No log	0.33	200	4.3357
No log	0.5	300	4.3897
No log	0.66	400	4.8238
4.4277	0.83	500	3.9088
4.4277	0.99	600	3.6063
4.4277	1.16	700	3.4278
4.4277	1.32	800	3.5428
4.4277	1.49	900	3.4331
3.0413	1.65	1000	3.3699
3.0413	1.82	1100	3.3622
3.0413	1.98	1200	3.5294
3.0413	2.15	1300	3.7918
3.0413	2.31	1400	3.4007
2.0843	2.48	1500	4.0296
2.0843	2.64	1600	4.1852
2.0843	2.81	1700	3.6690
2.0843	2.97	1800	3.6089
2.0843	3.14	1900	5.5534
1.7527	3.3	2000	4.7498
1.7527	3.47	2100	5.2691
1.7527	3.63	2200	5.1324
1.7527	3.8	2300	4.5912
1.7527	3.96	2400	4.1727
1.2037	4.13	2500	6.1174
1.2037	4.29	2600	5.7172
1.2037	4.46	2700	5.8843
1.2037	4.62	2800	6.4232
1.2037	4.79	2900	7.4486
0.8386	4.95	3000	7.1946
0.8386	5.12	3100	7.9869
0.8386	5.28	3200	8.0310
0.8386	5.45	3300	8.2954
0.8386	5.61	3400	8.5361
0.4389	5.78	3500	8.6040
0.4389	5.94	3600	8.5806

Featured Recommended AI Models

Empowering the Future, Your AI Solution Knowledge Base

English 简体中文繁體中文にほんご

Layoutlmv2 Large Uncased Finetuned Vi Infovqa

Model Overview

Model Features

Model Capabilities

Use Cases

🚀 layoutlmv2-large-uncased-finetuned-vi-infovqa

📄 License

📚 Documentation

Model description

Intended uses & limitations

Training and evaluation data

🔧 Technical Details

Training procedure

Training hyperparameters

Training results

Framework versions