Minh Nam TRAN

I have graduated from The University of Science, VNU-HCM, with a Bachelor of Science in Computer Science. I am supervised by Professor Dien Dinh and Dr. Long HB Nguen, where my thesis is titled “Exploring and Improving Language Understanding Abilities of Vietnamese Language Models.” Besides that, as an undergraduate research assistant at the Computational Linguistics Center, I study Vietnamese language models for natural language understanding tasks.

I am interested in Natural Language Processing, including (but not limited to):

Language Model-related Problems: Natural language understanding, logical reasoning, hallucination, knowledge integration, trustworthiness, and robustness.
Applications of Language Models: Using language models for domain-specific problems, real-world problems, or various NLP downstream tasks, such as information retrieval, question-answering, summarization, and dialogue systems.
Training/Learning Methods: Learning strategies for improving the performance of LLMs and how to make them think like humans.
Vision Language Model: Multimodal reasoning and understanding.

If you have any questions or want to collaborate, feel free to contact me via email.

Selected Publications

ViGLUE
ViGLUE: A Vietnamese General Language Understanding Benchmark and Analysis of Vietnamese Language Models

Minh-Nam Tran^*, Phu-Vinh Nguyen^*, Long Nguyen, and Dien Dinh

In Findings of the Association for Computational Linguistics: NAACL 2024 , Jun 2024

Abs Bib HTML PDF Code

As the number of language models has increased, various benchmarks have been suggested to assess the proficiency of the models in natural language understanding. However, there is a lack of such a benchmark in Vietnamese due to the difficulty in accessing natural language processing datasets or the scarcity of task-specific datasets. *ViGLUE*, the proposed dataset collection, is a Vietnamese General Language Understanding Evaluation benchmark developed using three methods: translating an existing benchmark, generating new corpora, and collecting available datasets. ViGLUE contains twelve tasks and encompasses over ten areas and subjects, enabling it to evaluate models comprehensively over a broad spectrum of aspects. Baseline models utilizing multilingual language models are also provided for all tasks in the proposed benchmarks. In addition, the study of the available Vietnamese large language models is conducted to explore the language models’ ability in the few-shot learning framework, leading to the exploration of the relationship between specific tasks and the number of shots.
@inproceedings{tran-etal-2024-viglue, title = {{V}i{GLUE}: A {V}ietnamese General Language Understanding Benchmark and Analysis of {V}ietnamese Language Models}, author = {Tran, Minh-Nam and Nguyen, Phu-Vinh and Nguyen, Long and Dinh, Dien}, editor = {Duh, Kevin and Gomez, Helena and Bethard, Steven}, booktitle = {Findings of the Association for Computational Linguistics: NAACL 2024}, month = jun, year = {2024}, address = {Mexico City, Mexico}, publisher = {Association for Computational Linguistics}, url = {https://aclanthology.org/2024.findings-naacl.261}, doi = {10.18653/v1/2024.findings-naacl.261}, pages = {4174--4189}, }
MultiView
Multi-perspective Traffic Video Description Model with Fine-grained Refinement Approach

Tuan-An To^*, Minh-Nam Tran^*, Trong-Bao Ho^*, Thien-Loc Ha^*, Quang-Tan Nguyen^*, Hoang-Chau Luong, Thanh-Duy Cao, and Minh-Triet Tran

In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops , Jun 2024

Abs Bib HTML PDF Code

The analysis of traffic patterns is crucial for enhancing safety and optimizing flow within urban cities. While urban cities possess extensive camera networks for monitoring the raw video data often lacks the contextual detail necessary for understanding complex traffic incidents and the behaviors of road users. This paper proposes a novel methodology for generating comprehensive descriptions of traffic scenarios combining a vision-language model (VLM) with rule-based refinements to capture pertinently pedestrian vehicle and environment factors. First a captioning model will generate a general description using processed video as input. Subsequently this description is refined sequentially through three primary modules: pedestrian-aware vehicle-aware and context-aware enhancing the final description. We evaluate our method on the Woven Traffic Safety datasets in Track 2 of the AI City Challenge 2024 obtaining competitive results with an S2 score of 22.6721. Code will be available at https://github.com/ToTuanAn/AICityChallenge2024_Track2
@inproceedings{To_2024_CVPR, author = {To, Tuan-An and Tran, Minh-Nam and Ho, Trong-Bao and Ha, Thien-Loc and Nguyen, Quang-Tan and Luong, Hoang-Chau and Cao, Thanh-Duy and Tran, Minh-Triet}, title = {Multi-perspective Traffic Video Description Model with Fine-grained Refinement Approach}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = jun, year = {2024}, pages = {7075-7084}, }