DIWA-Net Logo
DIWA-Net
About the project

DIWA-Net: A Parameter-Efficient Multi-Modal Architecture for Deepfake Detection in the Open-Set Paradigm

DIWA-Net is a research-grade multimodal deepfake detector that fuses DINOv2 visual features with Wav2Vec 2.0 audio representations through cross-modal attention and a balanced gated fusion head. The system is LoRA-adapted on the MAVOS-DD benchmark, supporting multiple languages and open-set evaluation against unseen manipulation methods.

Project details
ProgramFinal Year Project (FYP)
Academic year2026
DepartmentDepartment of Computer Science
UniversityUniversity of Engineering and Technology, Taxila
ModelDIWA-Net Phase 3 (LoRA)
BenchmarkMAVOS-DD · 12K training set
Authors

Student team

DIWA-Net was developed by a two-member student team over a full academic year at UET Taxila, under the supervision and co-authorship of Dr. Rabbia Mahum.

HA
Hamza Abdul Jabbar
Project Lead · Model & Backend
HN
Hamail Noor
Co-author · Research, Frontend & Documentation
Supervision

Supervisor & co-author

RM
Dr. Rabbia Mahum
Supervisor · Co-author · Supervision · Lecturer · Department of Computer Science

Dr. Rabbia Mahum supervised and co-authored this project — guiding problem formulation, model architecture, training strategy, experimental design, and thesis preparation at the Department of Computer Science, University of Engineering and Technology, Taxila.

Institution
University of Engineering and Technology, Taxila
Taxila, Punjab, Pakistan
www.uettaxila.edu.pk
Acknowledgements
Learn more

Contact: 22-CS-86@students.uettaxila.edu.pk