Most Video-Large Language Models (Video-LLMs) adopt an encoder-decoder framework, where a vision encoder extracts frame-wise features for processing by a language model. However, this approach incurs high computational costs, introduces resolution biases, and struggles to capture fine-grained multimodal interactions. To overcome these limitations, we propose ELVA, an encoder-free Video-LLM that directly models nuanced video-language interactions without relying on a vision encoder. ELVA employs token merging to construct a bottom-up hierarchical representation and incorporates a video guidance supervisor for direct spatiotemporal representation learning. Additionally, a hybrid-resolution mechanism strategically integrates high- and low-resolution frames as inputs to achieve an optimal balance between performance and efficiency. With only 7M publicly available video-text pairs, ELVA achieves performance on par with encoder-based Video-LLMs while reducing FLOPs by up to 95\% and inference latency by 92\%, offering a scalable and efficient solution for real-time video understanding.
View on arXiv@article{li2025_2503.18422,
title={ Breaking the Encoder Barrier for Seamless Video-Language Understanding },
author={ Handong Li and Yiyuan Zhang and Longteng Guo and Xiangyu Yue and Jing Liu },
journal={arXiv preprint arXiv:2503.18422},
year={ 2025 }
}