Back to Work & Research
Research
completed
2025
Vietnamese Fake News Detection (BiLSTM vs PhoBERT vs LLM)
Benchmarking classical deep learning and modern LLM prompting for Vietnamese news
Faculty-level research project on Vietnamese fake news detection using ReINTEL. Benchmarked BiLSTM, PhoBERT (frozen/fine-tuned), and LLM prompting; reported Accuracy/Macro-F1/AUC and performed error analysis and deployment trade-off review.
AI/ML
NLP
Research
Researcher
Developer
Timeline
2025
Type
Research
Status
completed
Outcome / Impact
- •Best model achieved Accuracy 0.963, Macro-F1 0.929, AUC 0.980 on test set
- •Produced benchmarking report with error analysis and inference trade-offs (latency/VRAM)
Tech / Skills
PyTorch
Transformers
PhoBERT
BiLSTM
Evaluation
Case Study
1) Context / Problem
What problem were you solving? Why does it matter?
2) Your Role
Responsibilities, decisions you owned, stakeholders.
3) Approach
Method, experiments, architecture, or research design.
4) Result / Impact
Metrics, rankings, adoption, business outcome.
5) Learnings
What you’d repeat or change next time.
6) Links
Add demos, repos, papers, slides, press, or media.