@inproceedings{liu-etal-2026-crisprune,
    title = "{C}ris{P}rune: Combining Contextual Relevance and Intrinsic Saliency for Efficient Visual Token Pruning in {MLLM}s",
    author = "Liu, Ziniu  and
      Zhou, Shuheng  and
      Liu, Mingqing  and
      Deng, Hao  and
      Zhu, Huijia",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Findings of the {A}ssociation for {C}omputational {L}inguistics: {ACL} 2026",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl-workshops/2026.findings-acl.663/",
    pages = "13546--13564",
    ISBN = "979-8-89176-395-1",
    abstract = "Visual token pruning has emerged as a pivotal strategy to alleviate the computational bottleneck in Multimodal Large Language Models (MLLMs), yet it frequently compromises the integrity of visual understanding in pursuit of efficiency. Existing methods face a fundamental tension: vision-centric approaches are susceptible to the attention sink phenomenon and operate in a query-agnostic manner, whereas text-guided methods often create an overly narrow focus, discarding essential background context and failing on ambiguous queries. In this paper, we propose CrisPrune, a training-free and model-agnostic method that reconciles efficiency with understanding by integrating visual saliency and text relevance. Specifically, we introduce intrinsic visual saliency with robust normalization to identify information-rich regions characterized by significant visual features. Simultaneously, we design dual-source text relevance to synergize explicit instruction alignment with implicit scene priors. Finally, we reformulate the selection process using a Determinantal Point Process (DPP) to balance token quality and spatial diversity. Extensive experiments demonstrate that CrisPrune significantly outperforms state-of-the-art methods. On LLaVA-NeXT, it achieves a 13 $\times$ decrease in FLOPs while maintaining 97{\%} of the original performance with 94.4{\%} of visual tokens pruned, effectively bridging the gap between efficiency and holistic understanding."
}