马普所&谷歌等提出AnyUp：无需再训练的通用特征上采样，推理即用，效果SOTA！
"视觉编码器" 相关文章
马普所&谷歌等提出AnyUp：无需再训练的通用特征上采样，推理即用，效果SOTA！
Data or Language Supervision: What Makes CLIP Better than DINO?
AndesVL Technical Report: An Efficient Mobile-side Multimodal Large Language Model
On Epistemic Uncertainty of Visual Tokens for Object Hallucinations in Large Vision-Language Models
On Epistemic Uncertainty of Visual Tokens for Object Hallucinations in Large Vision-Language Models
OpenVision 2：大道至简的生成式预训练视觉编码器
OpenVision 2：大道至简的生成式预训练视觉编码器
Improving Video Diffusion Transformer Training by Multi-Feature Fusion and Alignment from Self-Supervised Vision Encoders
LLaVA-MORE: A Comparative Study of LLMs and Visual Backbones for Enhanced Visual Instruction Tuning
Explaining How Visual, Textual and Multimodal Encoders Share Concepts
多模态系列-2：多视觉编码器协同与高低分辨率特征融合技术综述
Meta AI Introduces Perception Encoder: A Large-Scale Vision Encoder that Excels Across Several Vision Tasks for Images and Video
社区供稿 | “迷你版”书生·万象大模型开源，5% 参数实现九成性能，“小而美”易迁移
Apple Releases AIMv2: A Family of State-of-the-Art Open-Set Vision Encoders
多模态竞技场对标90B Llama 3.2，Pixtral 12B技术报告全公开
Microsoft Released LLM2CLIP: A New AI Technique in which a LLM Acts as a Teacher for CLIP’s Visual Encoder
谢赛宁新作：表征学习有多重要？一个操作刷新SOTA，DiT训练速度暴涨18倍
