TeX
0
2 commits
updated Jan 14, 2026
Authors: [Author Names] (HCMUS)
Competition: AbjadGenEval Shared Task 1 @ ArabicNLP 2025
Result: 1st Place (F1-Score: 0.93)
(Note: Refer to eacl2023.pdf for the vector TikZ diagram)
Our approach is detailed in our paper:
HCMUS_TheFangs at AbjadGenEval: Weighted Layer Pooling with Attention Fusion for Arabic AI-Generated Text Detection
pip install -r requirements.txt
Run the best performing experiment (exp19):
python exp/exp19_weighted_layer_pooling.py
We provide code for all 15+ approaches explored in our ablation study, including:
exp/exp04_discourse_motif.py: Discourse marker analysisexp/exp05_dediacritization.py: Diacritic-aware modelsexp/exp17_multi_contrastive.py: Contrastive Learning| Rank | Team | F1 Score | Accuracy | Precision | Recall |
|---|---|---|---|---|---|
| π₯ | HCMUS_TheFangs (Ours) | 0.93 | 0.93 | 0.97 | 0.89 |
| π₯ | chisboizhoigay | 0.90 | 0.91 | 0.95 | 0.86 |
| π₯ | alizain157 | 0.89 | 0.89 | 0.87 | 0.90 |
βββ exp/ # Experiment scripts (PyTorch/Transformers)
β βββ exp19_weighted... # β
WINNING SOLUTION
β βββ ... # Ablation studies
βββ EACL_2023.../ # LaTeX Source & PDF of the Paper
βββ training.csv # (Ignored) Training Data
βββ README.md # This file
TeX
0
2 commits
updated Jan 14, 2026
Authors: [Author Names] (HCMUS)
Competition: AbjadGenEval Shared Task 1 @ ArabicNLP 2025
Result: 1st Place (F1-Score: 0.93)
(Note: Refer to eacl2023.pdf for the vector TikZ diagram)
Our approach is detailed in our paper:
HCMUS_TheFangs at AbjadGenEval: Weighted Layer Pooling with Attention Fusion for Arabic AI-Generated Text Detection
pip install -r requirements.txt
Run the best performing experiment (exp19):
python exp/exp19_weighted_layer_pooling.py
We provide code for all 15+ approaches explored in our ablation study, including:
exp/exp04_discourse_motif.py: Discourse marker analysisexp/exp05_dediacritization.py: Diacritic-aware modelsexp/exp17_multi_contrastive.py: Contrastive Learning| Rank | Team | F1 Score | Accuracy | Precision | Recall |
|---|---|---|---|---|---|
| π₯ | HCMUS_TheFangs (Ours) | 0.93 | 0.93 | 0.97 | 0.89 |
| π₯ | chisboizhoigay | 0.90 | 0.91 | 0.95 | 0.86 |
| π₯ | alizain157 | 0.89 | 0.89 | 0.87 | 0.90 |
βββ exp/ # Experiment scripts (PyTorch/Transformers)
β βββ exp19_weighted... # β
WINNING SOLUTION
β βββ ... # Ablation studies
βββ EACL_2023.../ # LaTeX Source & PDF of the Paper
βββ training.csv # (Ignored) Training Data
βββ README.md # This file