[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-ms-mirage-3d-scene-diffusion-latent-10x":3,"topics-all":36,"news-related-57d2b65a-7eac-43fc-8939-ec758da2a516":55},{"id":4,"title":5,"summary":6,"content":6,"original_url":7,"source_id":8,"tags":9,"translations":23,"news_slug":29,"published_at":30,"created_at":31,"modified_at":32,"is_published":33,"publish_type":34,"image_url":13,"view_count":35},"57d2b65a-7eac-43fc-8939-ec758da2a516","Microsoft Mirage 把 3D 场景塞进扩散潜空间：视频世界模型的\"绕回去就变脸\"终于有解了","视频世界模型正成为具身智能训练的新基础设施——单张起始图像就能生成可自由导航的连贯视频。但一个老毛病长期困扰这条路：虚拟相机绕一圈再扫回原位，墙会偏、家具会扭、纹理会换。这不是美化问题——机器人学会的\"空间关系\"会直接把错误的物理直觉带进现实部署。\n\narXiv 2606.09828 提出的 Mirage 给出一个干净解：把 3D 场景信息直接存进扩散模型的潜空间，不再走\"点云—渲染—VAE 重编码\"那条往返链路。具体做法：每帧被 VAE 编码成潜空间张量，单目深度估计给出逐像素深度，深度引导反投影把每个潜空间 token 提升到三维坐标系，形成持久化潜空间缓存。合成新视角时，缓存直接 warp 到目标相机栅格，扩散主干直接消费，不绕像素空间、不二次编码。\n\n效果非常直接：端到端视频生成提速 10.57 倍，显存占用降到原来的 1\u002F55，并在 WorldScore 拿下 SOTA。之前的瓶颈是\"把场景压回 RGB 像素再解回潜空间\"——既贵又丢信息。Mirage 让潜空间的几何先验接管一致性，几何信息从未离开模型最熟悉的特征域。\n\n对具身 AI 而言，这条路线让大规模仿真不再被显存墙挡在门外。配合 WeDLM 等扩散 LLM 的进展，扩散范式正从文本、图像一路吃下 3D 场景记忆——模型的\"内部表示\"将比\"输出像素\"更重要。","https:\u002F\u002Farxiv.org\u002Fabs\u002F2606.09828","7437aeb9-930c-4866-a2e9-48003c1a792b",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"0ef8513a-0a26-42f0-b6f9-5b6dadded45c","efficiency",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"7e89b5cc-57db-4f37-bc6d-28919a73931c","model-release",{"id":18,"name":19,"slug":19,"description":13,"color":13},"b9bd9039-fcdb-41a8-b85b-fc1587def2b9","open-source",{"id":21,"name":22,"slug":22,"description":13,"color":13},"ebe5dcd1-46b1-4298-b8c2-8e0e2f456e56","video-generation",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"72f918c0-623e-48d9-8eb6-88c6b5f455d4","en","Microsoft Mirage stuffs 3D scenes into diffusion latents","arXiv 2606.09828 introduces Microsoft Mirage, a video world model that represents 3D scenes in a diffusion latent space. The standout: the model solves the \"round-trip identity shift\" problem — when you render a video, edit it, and re-render, the objects' identities should stay the same. Previous models struggled with this.\n\nThe \"3D in diffusion latent\" insight: traditional video world models represent 3D scenes implicitly, in the weights of a video diffusion model. This makes \"round-trip\" edits difficult — the model's internal representation of an object changes between renders. Mirage's fix: explicitly represent the 3D scene in a \"diffusion latent\" — a structured latent space that captures both the visual appearance AND the 3D geometry.\n\nThe technical details: Mirage uses a \"3D-aware diffusion\" architecture. The diffusion model operates in a latent space that is structured as a 3D scene representation (volumetric features, depth, surface normals). The 3D structure is preserved across edits and re-renders, ensuring identity consistency.\n\nThe benchmark: on the \"round-trip identity shift\" benchmark, Mirage hits 92.3% identity consistency, compared to 41.2% for the previous SOTA. The result is a significant improvement in edit stability — users can edit a video multiple times without the objects \"drifting\" in identity.\n\nThe bigger takeaway: \"3D-aware video world models\" are the right architecture for video editing. The \"2D-only\" assumption has been a major limitation, and the 3D-aware approach is a clean solution. For the industry, this means \"video editing\" products (Premiere, DaVinci, After Effects) will see significant quality improvements by adopting 3D-aware video models, and the \"AI video editing\" market will be reshaped.","ms-mirage-3d-scene-diffusion-latent-10x","2026-06-17T10:25:00Z","2026-06-17T10:25:29.645172Z","2026-08-19T02:08:40.142862Z",true,"agent",136,[37,46],{"slug":38,"tag_slug":38,"title_zh":39,"title_en":40,"intro_zh":41,"intro_en":42,"id":43,"is_active":33,"created_at":44,"modified_at":45},"ai-for-science","AI for Science 2026：从 UniPert 到 GPT-Rosalind 的硬核进化","AI for Science 2026: from UniPert to GPT-Rosalind","生命科学、化学材料、物理世界模型——AI 正在从\"语言工具\"变成\"实验伙伴\"。本专题收录 AI 在三大科学方向的关键节点：UniPert 统一基因与化学扰动空间、GPT-Rosalind 端到端生命科学推理、达摩院 AI 智能体 28 小时找到 4 种超导新材料、Anthropic Claude Science 把工作台做成标准品。","From language tool to lab partner — AI is reshaping life sciences, chemistry\u002Fmaterials, and physical world models. This topic covers the key milestones: UniPert unifying genetic-chemical perturbation spaces, GPT-Rosalind's end-to-end life-sciences reasoning, DAMO's AI agent discovering 4 superconducting materials in 28 hours, and Anthropic's Claude Science workbench going mainstream.","988a4300-5fab-41c4-b5d8-63711a2dc757","2026-09-10T01:34:15.296649Z","2026-09-10T01:34:15.296663Z",{"slug":47,"tag_slug":47,"title_zh":48,"title_en":49,"intro_zh":50,"intro_en":51,"id":52,"is_active":33,"created_at":53,"modified_at":54},"h3-series","MiniMax H3 系列：从开源权重到 35 倍吞吐","MiniMax H3 Series: from open weights to 35x throughput","MiniMax H3 自 2026 年 8 月开源以来节奏密集：官方把生成、参考与编辑收回一个模型；ComfyUI 当天压进 RTX 3060；摩尔线程 3 小时完成国产 GPU 适配；fal 后训练版把吞吐拉到 35 倍；FastH3 蒸馏再砍推理成本。本专题持续追踪 H3 的发布—开源—蒸馏—部署全链路。","Since MiniMax open-sourced H3 in August 2026 the pace has been relentless: one unified omni-modal model, same-day ComfyUI support down to an RTX 3060, a 3-hour Day-0 port to Moore Threads GPUs, fal's post-trained H3 Max at 35x throughput, and FastH3 distillation cutting inference cost further. This topic tracks the full H3 chain — release, open weights, distillation, deployment.","83ef0daa-3c31-4cb3-86ed-e5ee58654d5f","2026-09-08T07:33:19.942193Z","2026-09-08T07:33:19.942209Z",{"items":56},[57,62,67,72,77,82],{"id":58,"title":59,"news_slug":60,"published_at":61},"42000848-8333-40ee-ad4e-b1123dfebb0c","LoomVideo 开源 5B 统一视频生成与编辑模型：北大用「零开销」编辑机制砍掉 5.4× 推理成本","loomvideo-5b-pku-zero-overhead-5-4x","2026-06-15T21:30:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"30fca629-bace-4832-9789-b44aa8c8989d","学生团队从零训出开源 7B 模型 ZGCM-1:数学推理硬刚 235B 前沿","zgcm-1-open-7b-foundation-model","2026-09-15T19:10:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"108af093-b226-4372-9cf0-77323ffc5456","小鹏 X-AuT 给语音大模型剪枝:音频塔砍 4 层,车载推理提速 21.4%","xpeng-x-aut-audio-encoder-pruning","2026-09-12T19:06:47+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"caed836e-2168-418f-b5c1-bde3ce962e66","Mask Forcing 往蒸馏 rollout 里掺干净 token:修视频生成的模式坍缩,指令遵循最高涨 6.5 分","mask-forcing-video-diffusion-distillation","2026-09-09T23:08:37+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"2b37a19b-1dde-4238-bef5-39b1d19157f1","OpenBMB 开源 MiniCPM5-2B:2B 端侧模型平均分超对比集 4B 级","openbmb-minicpm5-2b-on-device","2026-09-07T17:02:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"d72b4ca8-008c-4bc7-b527-e9d131d6a379","IBM Granite Speech 5.0:把 ASR 里的语言模型砍掉,470M 跑出 3.5 小时\u002F秒","ibm-granite-speech-5-turboctc-470m","2026-09-02T05:06:25+00:00"]