5 papers
Retrieval Geometry Shapes Cache-Based Clip Adaptation
Mahir Shahriar Tamim, Md. Samiul Alim, Azmine Toushik Wasi +5
Cache-based test-time adaptation improves CLIP predictions by storing and retrieving examples from the target stream while keeping the model frozen. However, existing methods large…
Can LLMs Follow the Pulse of a Crisis? Evaluating Crisis Sentiment in Bangladesh's July Uprising
Md. Samiul Alim, Mahir Shahriar Tamim, Tanvir Ahmed Khan +4
Crisis sentiment analysis is especially challenging for low-resource languages such as Bangla, where language, context, and public reaction shift rapidly. We introduce UNRESTSENT20…
CAT-GS: Balanced Multimodal Learning via Calibrated Gating and Fusion Surgery
Mahir Shahriar Tamim, Sharjil Khan, Md. Samiul Alim +3
End-to-end training of multimodal neural networks often exhibits unstable neural dynamics characterized by three coupled failure modes that degrade learning: (i) modality imbalance…
When a Nation Speaks: Machine Learning and NLP in People's Sentiment Analysis During Bangladesh's 2024 Mass Uprising
Md. Samiul Alim, Mahir Shahriar Tamim, Maisha Rahman +2
Sentiment analysis, an emerging research area within natural language processing (NLP), has primarily been explored in contexts like elections and social media trends, but there re…
Teacher-Guided One-Shot Pruning via Context-Aware Knowledge Distillation
Md. Samiul Alim, Sharjil Khan, Amrijit Biswas +3
Unstructured pruning remains a powerful strategy for compressing deep neural networks, yet it often demands iterative train-prune-retrain cycles, resulting in significant computati…