Dense-Features on Xu'Bloghttps://xuquant.com/tags/dense-features/Recent content in Dense-Features on Xu'BlogXu'Bloghttps://xuquant.com/og-default.pnghttps://xuquant.com/og-default.pngHugo -- 0.152.2zhSat, 24 Jan 2026 10:00:00 +0800DINOv3：自监督视觉基模的规模化困局与 Gram Anchoring 破局https://xuquant.com/posts/world-models/dinov3/Sat, 24 Jan 2026 10:00:00 +0800https://xuquant.com/posts/world-models/dinov3/DINOv3 核心贡献剖析：Gram anchoring 如何解决大规模自监督训练中 dense feature 退化的根本问题，7B 参数 SSL 模型的训练工程，以及它在深度估计和 3D 匹配上的突破意味着什么。V-JEPA 2.1: When Self-Supervised Vision Learns to See Every Pixelhttps://xuquant.com/posts/world-models/vjepa-2.1/Sat, 10 Jan 2026 10:00:00 +0800https://xuquant.com/posts/world-models/vjepa-2.1/A deep analysis of V-JEPA 2.1's architectural innovations — dense predictive loss, deep self-supervision, multi-modal tokenizer, and scaling — tracing the path from collapsed context tokens to dense features that encode spatial structure, and the connection to depth estimation as geometric grounding.