PulseAugur
EN
LIVE 08:08:27
中文(ZH) RoboScience机器科学发布Visics通用具身大模型，实现跨本体、跨物体、跨任务｜最前线

RoboScience launches Visics embodied AI model with VLOA architecture

By PulseAugur Editorial · [3 sources] ·

RoboScience, an embodied intelligence company, has launched its Visics general-purpose embodied AI model. The model features a novel VLOA (Vision-Language-Object-Action) architecture designed to decouple cognition from execution by using Object Trajectory as an intermediate interface. This architecture aims to achieve broad generalization across different robots, objects, and tasks, demonstrated by its ability to perform complex operations like furniture assembly with high precision. AI

IMPACT This model's VLOA architecture and focus on generalization could accelerate the development of more versatile and capable robots across various industries.

RANK_REASON Frontier-lab model release with system card

Read on 36氪 (36Kr) →

AI-generated summary · Google Gemini · from 3 sources. How we write summaries →

RoboScience launches Visics embodied AI model with VLOA architecture

COVERAGE [3]

  1. 雷峰网 (Leiphone) TIER_1 中文(ZH) ·

    RoboScience releases general embodied large model Visics, fully demonstrating the VLOA dual-engine architecture for the first time

    <p style="text-align: left; margin-top: 6pt; margin-bottom: 6pt;">雷峰网获悉，<span style="font-family: Arial; font-size: 11pt;">6月24日，通用具身智能企业RoboScience机器科学在深圳成功举办通用具身大模型发布会。会上，公司首次完整披露了自研Visics大模型的技术架构VLOA（Vision-Language-Object-Action），并展示了模型在多项真实场景的应用，包括全球最复杂具身操作任务家具拼装。</span></p>…

  2. 36氪 (36Kr) TIER_1 中文(ZH) ·

    RoboScience releases Visics general embodied large model, achieving cross-ontology, cross-object, cross-task | Frontline

    <p>作者｜黄楠</p> <p>编辑｜袁斯来</p> <p>6月24日，通用具身智能企业RoboScience机器科学通用具身大模型发布，首次完整披露自研Visics大模型的技术架构VLOA（Vision-Language-Object-Action），并展示了模型在家具拼装、灵巧抓取、动态流水线等多项真实场景的应用。</p> <p>大语言模型有标准的文本Token，自动驾驶有统一的视觉或点云表征，这些基础格式的确定，让数据和模型可以在不同场景之间复用。但具身智能至今没有一个被行业公认的基础表征单元，其决定了数据如何采、模型从哪学、以及学习后是否能迁移至…

  3. Pandaily TIER_1 (CA) · [email protected] (Pandaily) ·

    RoboScience Unveils Visics, a General-Purpose Embodied AI Model

    Beijing-based embodied intelligence company RoboScience officially unveiled its general-purpose embodied AI model Visics on June 24, complete with a full techni...

RELATED ENTITIES

RELATED TOPICS