Tag
Hoover Institution's 'Only in America' interview with Fei-Fei Li — the 'Godmother of AI' — covering her journey to computer science, ImageNet's role in the deep learning revolution, and how World Labs' 'world models' will extend generative AI into spatial intelligence, plus discussions on truth-seeking and U.S. AI preparedness.
该论文通过设计科学研究方法，将 ANASSA 空间智能体 AI 编排框架从高层架构细化为包含 47 个逻辑角色、版本化任务图、类型化契约和治理控制的低层可执行架构，为受治理的空间智能体工作流提供实现级桥接。
MJev-Spatial 提出一种 System 1 多模态决策模型，将空间推理任务统一为单次前向的分类框架，利用预训练的 query/key 投影为候选选项打分，在 12 项基准中以 65.64% 取得最高总分，并比自回归基线快 2.5 至 5.6 倍。
本文提出 Mine Odyssey，一个基于 Minecraft 真实世界地点重建的 agentic spatial intelligence 基准，包含 20 个国家/地区的 30 个地点、180 项导航任务。评测显示 GPT-6 Astra 以 85.6% 成功率领先，Claude Opus 5.5 为 73.9%，而最强开源模型 DeepSeek-V4.1-Flash 仅 23.9%，凸显当前模型在空间智能上的差距。
AMD's deal with World Labs could influence future chip design by integrating hardware development with spatial AI research led by Fei-Fei Li, emphasizing closer collaboration between model builders and chip designers.
AMD is acquiring Fei-Fei Li's spatial-intelligence startup World Labs for $8.2 billion, and Li will join AMD as executive vice president and chief scientist.
Fei-Fei Li announces that her AI startup World Labs is joining AMD, highlighting the company's work in spatial intelligence and the release of Atlas, a model for new camera view prediction.
World model AI companies like AMI Labs and World Labs are maintaining secrecy about their commercialization plans due to competitive pressures and ease of fundraising in the industry.
VABench introduces a benchmark to evaluate embodied spatial intelligence in models by testing their ability to observe, reason, and act through visual demonstrations and active perception. It shows that active camera control improves task success, but no model completes long-horizon episodes.
This paper introduces SpatialBlock-15k, a synthetic dataset for block-stacking problems, to enhance 3D spatial reasoning in large vision-language models, demonstrating improved performance and generalization to real-world tasks.
Atlas is introduced as a foundation model for spatial intelligence in a video that features atmospheric music and sparse spoken words.
World Labs announced Atlas, an omni world model that natively processes text, images, video, and 3D geometry in a shared spatial context to advance AI capabilities in simulating space, time, and physical interactions.
World Labs introduces Atlas, a multimodal world model that can generate, reconstruct, and simulate 3D environments with camera control and spatial consistency.
LightNav-0 is a compact generalist navigation model that leverages pretrained vision-language models' spatial intelligence to achieve state-of-the-art embodied navigation across diverse tasks and robot embodiments.
This paper introduces a continuous metric field framework trained by a single causal contrastive loss that unifies geometric structure discovery from robot navigation to black hole emergence, demonstrating zero-shot generalization across dimensions.
Stanford HAI discusses the emergence of world models and spatial intelligence in AI, calling for governance frameworks that address capabilities beyond language processing.
World Labs shares early results on building worlds to train robots, advancing spatial intelligence for interaction with physical and virtual environments.
Promotes the eBook 'Python for GIS & Spatial Intelligence' which teaches interactive mapping and spatial analysis using Python.
Google Maps Platform Product Manager Angela Yu discussed the shift from developers to builders in the AI era at NYC Community Day, emphasizing agents, spatial intelligence, and how Google's data empowers builders.
An Ars Technica article explores the promise and limitations of world models as an emerging AI paradigm, contrasting them with LLMs and featuring expert insights on their applications in robotics, research, and asset generation.