Imran Razzak

4

Papers

22

Total Citations

Papers (4)

Seeing Far and Clearly: Mitigating Hallucinations in MLLMs with Attention Causal Decoding

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

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

NeurIPS 2025arXiv

DHGCN: Dynamic Hop Graph Convolution Network for Self-Supervised Point Cloud Learning