作者位次 #2/5ACL 2026Poster· Natural Language Processing
HERMES : KV Cache as Hierarchical Memory for Efficient Streaming Video Understanding
Haowei Zhang · Shudong Yang · Jinlan Fu · See-Kiong Ng · Xipeng Qiu
摘要
Recent advancements in Multimodal Large Language Models (MLLMs) have demonstrated significant improvement in offline video understanding. However, extending these capabilities to streaming video inputs, remains challenging, as existing models struggle to simultaneously maintain stable understandi...