[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"news-slug-deeproute-large-model-autonomous-driving-ruanchong":3,"topics-all":36,"news-related-4c1a894b-97fc-4d96-890c-8888167e23d8":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},"4c1a894b-97fc-4d96-890c-8888167e23d8","元戎启行全面押注大模型：自动驾驶路线彻底转向","在2026北京车展上，前DeepSeek多模态技术核心研究员阮翀以元戎启行首席科学家身份首次公开亮相，随之而来的是该公司技术路线的根本性转变——全面押注大模型自动驾驶。\n\n元戎启行CEO周光表示，2026年初多模态大模型能力取得突破性进展，大模型自动驾驶路线的起点已远优于上一代技术。这意味着行业正式承认了一个事实：靠堆砌小模型打天下的时代正在终结。\n\n元戎启行明确指出了小模型路线的核心问题——「跷跷板效应」：小模型在不同场景间切换时会出现能力折返，解决了一个corner case，却可能在另一个场景降级。这不是算法调优能彻底解决的问题，而是架构层面的先天缺陷。全场景安全覆盖，需要的是更强的泛化能力。\n\n新架构下，元戎启行正从多个小模型向统一基座大模型迁移，分化为驾驶、分析、评论三个垂直模型。真正的亮点在于迭代效率的跃升：单次模型迭代周期从100余小时压缩至10余小时，快了近10倍。迭代周期缩短意味着路测发现的问题能快速修复，Corner Case的覆盖速度大幅提升，直接决定了技术收敛的速度。\n\n这不是一家公司的路线选择，而是整个自动驾驶行业的风向标。当多模态大模型能力跨过临界点，「大模型为主、小模型做保障」的混合架构将逐步成为主流。阮翀的加入意味着DeepSeek在多模态领域的积累正在向自动驾驶领域溢出，大模型竞争正在进入跨界整合的新阶段。","https:\u002F\u002Fegs.stcn.com\u002Fnews\u002Fdetail\u002F2279542.html","7a2a9955-6324-4889-a5cc-fd2c967dd418",[10,14,17,20],{"id":11,"name":12,"slug":12,"description":13,"color":13},"e676a5cf-1f24-472f-a765-86fa21a1bc3c","ai-model",null,{"id":15,"name":16,"slug":16,"description":13,"color":13},"a8002d98-9df1-4ab9-94d4-a7625af634c4","china-ai",{"id":18,"name":19,"slug":19,"description":13,"color":13},"01598627-1ea6-4b27-a5d8-874971571a71","llm",{"id":21,"name":22,"slug":22,"description":13,"color":13},"499f4b56-819d-49a3-9609-33e775143b86","multimodal",[24],{"id":25,"lang":26,"title":27,"summary":28,"content":13},"2832072d-27cf-4222-b99e-5638ebb73690","en","DeepRoute.ai goes all-in on foundation models for autonomy","At the 2026 Beijing Auto Show, former DeepSeek multimodal technology core researcher Ruan Chong made his first public appearance as Chief Scientist at DeepRoute.ai, accompanied by a fundamental shift in the company's technology route — a full bet on large-model autonomous driving.\n\nDeepRoute.ai CEO Zhou Guang said multimodal large model capability achieved breakthrough progress in early 2026, and the starting point of the large-model autonomous-driving route is already far superior to the previous-generation technology. This means the industry has officially acknowledged a fact: the era of stacking small models is ending.\n\nDeepRoute.ai clearly pointed out the core problem of the small-model route — the \"seesaw effect\": when small models switch between different scenarios, capability will rebound, solving one corner case may degrade in another. This isn't a problem algorithm tuning can completely solve, but a fundamental architectural defect. Full-scenario safety coverage requires stronger generalization capability.\n\nUnder the new architecture, DeepRoute.ai is migrating from multiple small models to a unified base large model, splitting into three vertical models: driving, analysis, and commentary. The real highlight is the iteration efficiency leap: single model iteration cycle compressed from 100+ hours to 10+ hours, nearly 10× faster. Shortened iteration cycle means road-test-discovered problems can be fixed quickly, corner case coverage speed greatly increases, directly determining the speed of technology convergence.\n\nThis isn't one company's route choice, but a wind vane for the entire autonomous-driving industry. When multimodal large model capability crosses the critical point, the \"main large model, small model as guarantee\" hybrid architecture will gradually become mainstream. Ruan Chong's joining means DeepSeek's accumulation in multimodal is spilling over into autonomous driving, and large-model competition is entering a new stage of cross-domain integration.","deeproute-large-model-autonomous-driving-ruanchong","2026-04-26T22:01:00Z","2026-04-26T22:04:41.995252Z","2026-08-19T02:08:40.142862Z",true,"agent",182,[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},"3b7833c1-4213-4155-9125-df436adf96d7","大模型机器人的空间盲区被攻破：RAM模型让机器人真正看懂三维世界","ram-zhejiang-cuhk-3d-spatial-science-robotics","2026-05-06T13:00:00+00:00",{"id":63,"title":64,"news_slug":65,"published_at":66},"2fc64783-8b2a-49a3-939b-edf02bff3622","Ox Alpha 指纹指向 GLM-5.3:OpenRouter 的 1M 上下文隐身模型可能是智谱","ox-alpha-glm-5-3-stealth-zhipu","2026-08-22T14:00:00+00:00",{"id":68,"title":69,"news_slug":70,"published_at":71},"ad10985b-425c-4af1-9495-c63792a2b593","腾讯混元把语音识别打到 3% WER：Hy ASR 3.0 preview 让 ASR 从“逐字”走向“读语境”","tencent-hunyuan-hy-asr-3-0-preview-context-aware","2026-08-05T00:00:00+00:00",{"id":73,"title":74,"news_slug":75,"published_at":76},"a151db0c-d832-4df2-ac03-2d4e58b26e99","Kimi K3 跑通 MiniTriton:Moonshot 让 LLM 第一次从零编译出自己的 GPU 编译器","kimi-k3-minitriton-gpu-compiler","2026-07-26T14:00:00+00:00",{"id":78,"title":79,"news_slug":80,"published_at":81},"bd050dd6-4d85-4616-a004-55c23c533a24","腾讯混元合并大语言模型与多模态团队，成立基础模型部探索全模态统一","tencent-hunyuan-foundation-model-dept","2026-07-24T03:00:00+00:00",{"id":83,"title":84,"news_slug":85,"published_at":86},"2ada2e69-25c2-4951-9e49-2b24a043393e","腾讯 Marvis 把 Agent 拽到端侧:混元要做 PC 集群,应用宝做了「系统级」分诊","tencent-marvis-on-device-agent","2026-07-23T20:30:00+00:00"]