作者位次 #6/13ACL 2026Poster· Natural Language Processing
Attn- GS : Attention-Guided Context Compression for Efficient Personalized LLM s
Shenglai Zeng · Tianqi Zheng · Chuan Tian · Dante Everaert · Yau-Shian Wang · Yupin Huang · Michael J. Morais · Rohit Patki · Jinjin Tian · Xinnan Dai · Kai Guo · Monica Xiao Cheng · Hui Liu
摘要
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...