一作NeurIPS 2026Poster· Computer Vision / Vision Models & Multimodal
Video-FLAIR: Not Whether to Reason, But How
Yogesh Kulkarni · Pooyan Fazli
首次获取该候选人的完整档案及 StartLux 评估共 20 积分；每位用户单独解锁，之后查看免费。主动刷新 20 积分。
实时查找电子邮箱，费用不超过20积分；未找到邮箱不扣费。
Yogesh Kulkarni · Pooyan Fazli
Yogesh Kulkarni · Pooyan Fazli
Chaoyu Li · Yogesh Kulkarni · Pooyan Fazli
The computational cost of training multimodal large language models (MLLMs) grows rapidly with the number of processed tokens. Existing efficiency methods mainly target inference via token reduction or merging, offering limited benefits during training. We introduce ReGATE ( Re ference- G uided A...