Skip to main content

VLAI Documents

Join VLAI - Never SAI

Research Areas

💬

Vi Natural Language Processing

Research focused on Vietnamese natural language processing tasks, including text classification, sentiment analysis, and question answering.

  • ViVQA
🧠

Vi Large Language Models

Development and fine-tuning of large language models specifically for the Vietnamese language, aiming to improve performance on various NLP benchmarks.

  • ViLLM
🎤

Vi Speech and Audio Processing

Investigation into speech recognition, synthesis, and audio processing techniques tailored for Vietnamese, enhancing voice-based applications.

  • ViSAP
🤖

Vi Agents and Applications

Exploration of intelligent agents and their applications in Vietnamese contexts, including chatbots, virtual assistants, and automated customer service solutions.

  • ViAA

Our Team

Duong Dinh Thang

Duong Dinh Thang

Researcher

Research interests in AI and Machine Learning

Duong Nguyen Thuan

Duong Nguyen Thuan

Researcher

Research interests in AI and Machine Learning

Duong Truong Bình

Duong Truong Bình

Researcher

Research interests in AI and Machine Learning

Nguyen Anh Khoi

Nguyen Anh Khoi

Researcher

Research interests in AI and Machine Learning

Publications

Heterogeneous Transfer Learning Using Pre-trained Feature Mapping and Exchange

Nguyen Thuan Duong, Dinh Thang Duong, Hong Phuc Nguyen, Quang Vinh Dinh

AICI 2024 (2024)

This paper introduces novel methods for addressing transfer learning across different model architectures.

deep-learningtransfer learning

Enhancing Vietnamese VQA through Curriculum Learning on Raw and Augmented Text Representations

khoi_anh_nguyen, linh_yen_vu, thang_dinh_duong, thuan_nguyen_duong, huy_thanh_nguyen, vinh_quang_dinh

AAAI Workshop 2025 (2024)

We propose a Vietnamese VQA framework combining paraphrase-based augmentation and dynamic curriculum learning to enhance model generalization and performance in low-resource settings.

vqaaugmentationcurriculum learning

An Automate Pipeline for Contructing a Vietnamese VQA-NLE Dataset

Duong Truong Binh, Duong Dinh Thang

ICISN 2025 (2025)

ViVQA-X is a Vietnamese adaptation of the VQA-X dataset for VQA-NLE tasks, featuring translated and quality-evaluated question–answer–explanation pairs aligned with COCO images.

datasetvqaXAI
Easy to Use

Comprehensive Tutorials

Learn MLOps, DevOps, and AI best practices through our detailed tutorials and step-by-step guides covering the latest tools and technologies.

Focus on What Matters

Practical Knowledge

Focus on real-world applications with hands-on examples and code snippets. Everything you need to deploy AI systems in production.

Powered by React

Community Driven

Join our community of researchers and practitioners. Share knowledge, collaborate on projects, and advance AI research together.