作者位次 #3/11ACL 2026Poster· Natural Language Processing
GMSA : Enhancing Context Compression via Group Merging and Layer Semantic Alignment
Jiwei Tang · ZHICHENG ZHANG · Shunlong Wu · Jingheng Ye · Lichen Bai · Zitai Wang · Tingwei Lu · Lin Hai · Yiming Zhao · Hai-Tao Zheng · Hong-Gee Kim
摘要
Large Language Models (LLMs) have achieved remarkable performance across a wide range of Natural Language Processing (NLP) tasks. However, in long-context scenarios, they face two challenges: high computational cost and information redundancy. To address these challenges, we propose GMSA, an enco...