浪费时间
发表于 2025-3-23 11:11:05
,PromptCCD: Learning Gaussian Mixture Prompt Pool for Continual Category Discovery, data while mitigating the challenge of catastrophic forgetting—an open problem that persists even in conventional, fully supervised continual learning. To address this challenge, we propose PromptCCD, a simple yet effective framework that utilizes a Gaussian Mixture Model (GMM) as a prompting metho
irreparable
发表于 2025-3-23 14:45:59
http://reply.papertrans.cn/25/2424/242308/242308_12.png
floodgate
发表于 2025-3-23 21:14:39
http://reply.papertrans.cn/25/2424/242308/242308_13.png
镇压
发表于 2025-3-23 22:32:58
http://reply.papertrans.cn/25/2424/242308/242308_14.png
搬运工
发表于 2025-3-24 05:41:50
,NOVUM: Neural Object Volumes for Robust Object Classification,ects. In this work, we show that explicitly integrating 3D compositional object representations into deep networks for image classification leads to a largely enhanced generalization in out-of-distribution scenarios. In particular, we introduce a novel architecture, referred to as NOVUM, that consis
harbinger
发表于 2025-3-24 06:59:53
http://reply.papertrans.cn/25/2424/242308/242308_16.png
山崩
发表于 2025-3-24 11:55:46
http://reply.papertrans.cn/25/2424/242308/242308_17.png
Interferons
发表于 2025-3-24 15:53:10
http://reply.papertrans.cn/25/2424/242308/242308_18.png
整体
发表于 2025-3-24 20:17:49
,ColorMNet: A Memory-Based Deep Spatial-Temporal Feature Propagation Network for Video Colorization,sion or recurrently propagating estimated features that will accumulate errors or cannot explore information from far-apart frames, we develop a memory-based feature propagation module that can establish reliable connections with features from far-apart frames and alleviate the influence of inaccura
鸟笼
发表于 2025-3-25 02:29:03
http://reply.papertrans.cn/25/2424/242308/242308_20.png