5 papers
ReCoVR: Closing the Loop in Interactive Composed Video Retrieval
Bingqing Zhang, Yi Zhang, Zhuo Cao +4
Composed video retrieval (CoVR) searches for target videos using a reference video and a modification text, but existing methods are restricted to a single interaction round and ca…
Robust Test-time Video-Text Retrieval: Benchmarking and Adapting for Query Shifts
Bingqing Zhang, Zhuo Cao, Heming Du +4
Modern video-text retrieval (VTR) models excel on in-distribution benchmarks but are highly vulnerable to real-world query shifts, where the distribution of query data deviates fro…
RePaViT: Scalable Vision Transformer Acceleration via Structural Reparameterization on Feedforward Network Layers
Xuwei Xu, Yang Li, Yudong Chen +2
We reveal that feedforward network (FFN) layers, rather than attention layers, are the primary contributors to Vision Transformer (ViT) inference latency, with their impact signify…
ROLeR: Effective Reward Shaping in Offline Reinforcement Learning for Recommender Systems
Yi Zhang, Ruihong Qiu, Jiajun Liu +1
Offline reinforcement learning (RL) is an effective tool for real-world recommender systems with its capacity to model the dynamic interest of users and its interactive nature. Mos…
General Scene Adaptation for Vision-and-Language Navigation
Haodong Hong, Yanyuan Qiao, Sen Wang +2
Vision-and-Language Navigation (VLN) tasks mainly evaluate agents based on one-time execution of individual instructions across multiple environments, aiming to develop agents capa…