← Search

Zongpu Zhang

2 accepted papers

2026

SegMo: Co-Designing Content-Aware Sparsity and Locally-Cohesive Segment Parallelism for Efficient VLM Inference

CVPR 2026

Video Large Language Models (VideoLLMs) face a fundamental performance bottleneck: the token explosion intrinsic to video inputs. The resulting O(N^2) prefill cost makes conventional Transformer inference prohibitively expensive at scale. Existing attempts fall into a hard accuracy-latency dilemma:

Cited by 0SourcecodeScholar
2019

Object Guided External Memory Network for Video Object Detection

ICCV 2019poster

Video object detection is more challenging than image object detection because of the deteriorated frame quality. To enhance the feature representation, state-of-the-art methods propagate temporal information into the deteriorated frame by aligning and aggregating entire feature maps from multiple n…

Cited by 137PDFScholar