PhAI Labs and universities unveil JEPA-Anything, a cross-domain framework for learning predictive models across seven systems
PhAI Labs, together with researchers from Oxford, Stanford, Princeton, and CUHK, released JEPA-Anything, a cross-domain framework that learns predictive models for different 'worlds' by sharing a prediction core and a unified latent world-state interface while keeping domain-specific encoders. The framework was validated across seven system types, including patient-derived organoids and simulated planetary orbits, where the learned orbital frequency–semimajor axis relation matched Kepler's third law with high precision. The work suggests a common underlying method for AI to understand change across diverse domains.
Coverage timeline
量子位量子位
AI离“理解万物”还有多远？先拿癌细胞和行星轨道试试水 梦瑶 2026-09-19 14:39:32 来源： 量子位 同一预测核心，跨七类系统验证 衡宇 发自 凹非寺 量子位 | 公众号 QbitAI 一篇技术报告中居然出现了两个看起来风马牛不相及的研究对象，跟着AI真是见世面。 一个是 患者来源的类器官 ，另一个则 跨越到了宏观宇宙 …… 先说前者。研究者利用模型学到的内部因子，提出了细胞因子联合抗体阻断的干预方案，并经细胞系、类器官、小鼠实验层层验证。 至于后者，研究者把模拟的行星位置和速度轨迹喂给模型，不告诉它开普勒定律。 等它学完后分析其内部的预测模式，提取出的“轨道频率-半长轴”关系拟合斜率达-1.4991（开普勒第三定律理论值为-1.5），且R²=0.99999999。 肿瘤研究和行星轨道计算 两个八竿子打不着的领域，居然能用同一种预测学习方法来搞定。 大千世界千差万别，天气、分子、细胞、病人、机器人乃至行星，各自遵循完全不同的规律。 但 AI在理解它们“如何变化”时，有没有可能共享一套更底层的方法？ 最近，PhAI Labs联合牛津、斯坦福、普林斯顿、港中文等高校团队最新发布的 JEPA-Anything ，就在尝试回答这个问题。 论文显示，JEPA-Anything是一个 面向不同“世界”学习预测模型的跨领域框架 。 每个领域保留自己的观测形式、context-target构造和编码器，但共享后面的预测核心以及一套统一的latent world-state interface。 其中涉及一个叫做 OPF （Orthogonal Predictive Factorization，正交预测分解） 的关键机制 。 可以先简单理解为： 别把世间各种各样的变化，全都堆在一块去学习建模。 给世界建模，一条预测通路可能不够 首先来厘清一个概念，在JEPA-Anything这项研究中，“世界模型”究竟意味着什么？ JEPA-Anything论文给出的定义兼具宽泛度与操作性。 论文写道， 只要模型能根据现有信息 （Context） 建立内部状态，并以此预测同一系统中的另一个状态，均可归入隐空间世界模型 （Latent World Model） 的范畴。 这里的“另一个状态”，既可以是时间上的未来，也可以是空间上的隐藏区域、不同视角，甚至是施加某种干预后的结果。 顺着这
Related stories
Tadpole: Autoencoder Foundation Model for 3D PDEs with Online Learning
Researchers propose Tadpole, an autoencoder-based foundation model for 3D partial differential equations (PDEs), which learns to compress physical states before predicting evolution, with online learning capabilities. The approach differs from typical predictive models by first learning the spatial structure of physical fields, aiming to improve transferability across different physical systems.
1 reports · 1 sources
Blinded benchmark of 511 AI-designed antibodies finds models struggle on unknown targets
A blinded prospective benchmark published in Nature Biotechnology on August 16, 2026 tested 511 AI-designed or predicted antibodies from 29 organizations against SARS-CoV-2 RBD. The results indicate that current AI models struggle to reliably generate high-affinity, developable antibodies for unknown targets, even for a well-studied antigen. The study highlights the gap between retrospective performance and prospective utility in antibody discovery.
OpenAI launches GPT-6 Astra, claims world's best computer use model
OpenAI announced GPT-6 Astra, its next-generation model focused on computer and browser use, claiming it is the world's most intelligent and aligned model. The company stated Astra was built on its largest training run using over 100,000 GPUs at the Stargate facility, and in tests it outperformed humans in tasks like booking DMV appointments and searching job listings. OpenAI president Greg Brockman declared 'Welcome to the AGI era,' though the model's capabilities and limitations are still being assessed by developers.
23 reports · 9 sources515 points · 162 comments
Z.ai confirms Ox Alpha is GLM-5.3 Flash, releases open weights
Z.ai confirmed that the anonymous model 'Ox Alpha', which went viral on OpenRouter, is a new iteration of its GLM series, specifically GLM-5.3 Flash, and released its weights on Hugging Face under a new license. The model, a 320B-parameter natively multimodal model running on domestic chips, topped OpenRouter's leaderboard and ended DeepSeek's 56-day streak on OpenCode. The license requires companies with over $10B in revenue to pass a security review.
9 reports · 4 sources801 points · 281 comments
News updates every hour. Subscribe in your RSS reader to get new stories automatically.
Subscribe