Imran Razzak

CARL: Preserving Causal Structure in Representation Learning

PAL: Probing Audio Encoders via LLMs - Audio Information Transfer into LLMs

StreamAgent: Towards Anticipatory Agents for Streaming Video Understanding

Consistent Region-Informed Self-supervised Pretraining

Rolling Ball Optimizer: Learning by ironing out loss landscape wrinkles

C$^2$-AFCL: Cross-task Calibration for Asynchronous Federated Continual Learning

Hierarchical Sequence Iteration for Heterogeneous Question Answering

$S^3$-Bench: A Comprehensive Study of Multimodal LLMs for Scientific Discovery with Benchmarking

Decoding Causal Structure: End-to-End Mediation Pathways Inference

Towards Dynamic 3D Reconstruction of Hand-Instrument Interaction in Ophthalmic Surgery

CG-SSL: Concept-Guided Self-Supervised Learning

UniViT: Unifying Image and Video Understanding in One Vision Encoder

Dual-Stage Value-Guided Inference with Margin-Based Reward Adjustment for Fast and Faithful VLM Captioning

CALM: Culturally Self-Aware Language Models