English

LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs

Computer Vision and Pattern Recognition 2026-05-26 v2

Abstract

The fundamental challenge in scaling Video Large Language Models (Video LLMs) to long-form video lies in managing the explosion of visual-token context length. Existing strategies predominantly focus on "post-hoc" token reduction -- reducing visual tokens after feature extraction to alleviate the LLM's computational overhead. While these methods effectively reduce the number of visual tokens, we observe that the primary latency bottleneck then shifts from the LLM to the expensive per-frame processing of the vision encoder. To address this, we introduce LiteFrame, a strong, yet highly efficient video encoder backbone for Video LLMs. To train LiteFrame, we propose Compressed Token Distillation (CTD), a novel training framework that teaches a compact student vision encoder to directly predict information-dense, spatio-temporally compressed representations produced by a large teacher vision model, effectively bypassing redundant computation. When coupled with further Language Model Adaptation (LMA), this approach results in a new latency-accuracy Pareto frontier -- compared with InternVL3-8B, LiteFrame provides a 35% reduction in end-to-end latency while processing 8×\times more frames and improves average video understanding accuracy across multiple benchmarks. Our results demonstrate a new potential path to unlocking longer-form video understanding under fixed compute budgets.

Keywords

Cite

@article{arxiv.2605.17260,
  title  = {LiteFrame: Efficient Vision Encoders Unlock Frame Scaling in Video LLMs},
  author = {Jihwan Kim and Nikhil Parthasarathy and Danfeng Qin and Junhwa Hur and Deqing Sun and Bohyung Han and Ming-Hsuan Yang and Boqing Gong},
  journal= {arXiv preprint arXiv:2605.17260},
  year   = {2026}
}

Comments

Project Page: https://jjihwan.github.io/projects/LiteFrame

R2 v1 2026-07-22T07:17:04.631Z