作者位次 #5/5ACL 2026Poster· Natural Language Processing
Vista- LLM : Decoupled Query-Guided Visual Token Pruning for Efficient Long-Video Large Language Models
Zhenyu Li · Zuchao Li · Ping Wang · Lefei Zhang · Haojun Ai
摘要
Long-video understanding is bottlenecked by the high cost of processing massive visual tokens. Current reduction strategies often rely on static allocation or inefficient in-network selection that disrupts optimized attention kernels. In this paper, we introduce Vista-LLM, a decoupled framework f...