@inproceedings{lu-etal-2025-vipe,
    title = "{V}i{PE}: Visual Perception in Parameter Space for Efficient Video-Language Understanding",
    author = "Lu, Shichen  and
      Yue, Tongtian  and
      Guo, Longteng  and
      Li, Handong  and
      He, Xingjian  and
      Liu, Si  and
      Liu, Jing",
    editor = "Christodoulopoulos, Christos  and
      Chakraborty, Tanmoy  and
      Rose, Carolyn  and
      Peng, Violet",
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-emnlp/2025.emnlp-main.897/",
    pages = "17775--17786",
    ISBN = "979-8-89176-332-6",
    abstract = "Existing video-language models (Video-LLMs) typically rely on concatenating visual tokens with textual inputs for joint modeling. However, this token-level alignment leads to significant inefficiency, especially when scaling to long videos with dense visual inputs. In this work, we propose a video-to-parameter efficiency paradigm named ViPE that eliminates redundant visual tokens by transforming video content into visual perceptual weights, which are directly injected into the LLM{'}s parameters. ViPE consists of a visual injection module that compresses video features into a small set of perceptual queries using a hierarchical merge strategy, and a visual perception module that integrates the resulting representations into the LLM through a lightweight LoRA-like mechanism. ViPE achieves performance comparable to token-based baselines such as LLaVA, while reducing FLOPs by 85{\%} and inference time by up to 65{\%}, demonstrating a highly efficient and scalable solution for video understanding."
}