@inproceedings{tang-etal-2025-perception,
    title = "Perception Compressor: A Training-Free Prompt Compression Framework in Long Context Scenarios",
    author = "Tang, Jiwei  and
      Xu, Jin  and
      Lu, Tingwei  and
      Zhang, Zhicheng  and
      YimingZhao, YimingZhao  and
      LinHai, LinHai  and
      Zheng, Hai-Tao",
    editor = "Chiruzzo, Luis  and
      Ritter, Alan  and
      Wang, Lu",
    booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
    month = apr,
    year = "2025",
    address = "Albuquerque, New Mexico",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/Ingest-2025-COMPUTEL/2025.findings-naacl.229/",
    pages = "4093--4108",
    ISBN = "979-8-89176-195-7",
    abstract = "Large language models (LLMs) demonstrate exceptional capabilities in various scenarios. However, they suffer from much redundant information and are sensitive to the position of key information in long context scenarios. To address these challenges, we present Perception Compressor, a training-free prompt compression framework. It includes a perception retriever that leverages guiding questions and instruction to retrieve the most relevant demonstrations, a dual-slope ratio allocator to dynamically allocate compression ratios and open-book ratios, and a semi-guided iterative compression that retains key information at the token level while removing tokens that distract the LLM. We conduct extensive experiments on long context benchmarks, i.e., NaturalQuestions, LongBench, and MuSiQue. Experiment results show that Perception Compressor outperforms existing methods by a large margin, achieving state-of-the-art performance."
}