← Search

Pratik Prabhanjan Brahma

2 accepted papers

2026

DUET-VLM: Dual stage Unified Efficient Token reduction for VLM Training and Inference

CVPR 2026

Vision-language models (VLMs) have achieved remarkable multimodal understanding and reasoning capabilities, yet remain computationally expensive due to dense visual tokenization. Existing efficiency approaches either merge redundant visual tokens or drop them progressively in language backbone, ofte

Cited by 0SourcecodeScholar
2019

Analyzing the Variety Loss in the Context of Probabilistic Trajectory Prediction

ICCV 2019poster

Trajectory or behavior prediction of traffic agents is an important component of autonomous driving and robot planning in general. It can be framed as a probabilistic future sequence generation problem and recent literature has studied the applicability of generative models in this context. The vari…

Cited by 71PDFcodeScholar