@inproceedings{zeng-etal-2026-attn,
    title = "Attn-{GS}: Attention-Guided Context Compression for Efficient Personalized {LLM}s",
    author = "Zeng, Shenglai  and
      Zheng, Tianqi  and
      Tian, Chuan  and
      Everaert, Dante  and
      Wang, Yau-Shian  and
      Huang, Yupin  and
      Morais, Michael J.  and
      Patki, Rohit  and
      Tian, Jinjin  and
      Dai, Xinnan  and
      Guo, Kai  and
      Cheng, Monica Xiao  and
      Liu, Hui",
    editor = "Liakata, Maria  and
      Moreira, Viviane P.  and
      Zhang, Jiajun  and
      Jurgens, David",
    booktitle = "Proceedings of the 64th Annual Meeting of the {A}ssociation for {C}omputational {L}inguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2026",
    address = "San Diego, California, United States",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/ingest-acl/2026.acl-long.1663/",
    pages = "35925--35946",
    ISBN = "979-8-89176-390-6",
    abstract = "Personalizing large language models (LLMs) to individual users requires incorporating extensive interaction histories and profiles, but input token constraints make this impractical due to high inference latency and API costs. Existing approaches rely on heuristic methods such as selecting recent interactions or prompting summarization models to compress user profiles. However, these methods treat context as a monolithic whole and fail to consider how LLMs internally process and prioritize different profile components. We investigate whether LLMs' attention patterns can effectively identify important personalization signals for intelligent context compression. Through preliminary studies on representative personalization tasks, we discover that (a) LLMs' attention patterns naturally reveal important signals, and (b) fine-tuning enhances LLMs' ability to distinguish between relevant and irrelevant information. Based on these insights, we propose \textbf{Attn-GS}, an attention-guided context compression framework that leverages attention feedback from a marking model to mark important personalization sentences, then guides a compression model to generate task-relevant, high-quality compressed user contexts. Extensive experiments demonstrate that Attn-GS significantly outperforms various baselines across different tasks, token limits, and settings, achieving performance close to using full context while reducing token usage by 50 times."
}