腾讯网

RoboChallenge榜首模型开源

AIPress.com.cn报道1月12日消息，具身人工智能初创公司 Spirit AI 宣布，其最新一代视觉-语言-动作（Vision-Language-Action，VLA）模型 Spirit v1.5 在 RoboChallenge 基准测试中获得综合排名第一同时，将模型权重、核心代码及评测流程全面开源，以支持结果复现与学术验证。RoboChallenge 是一个面向真实机器人执行场景的..
雷锋网

一文纵览 Vision-and-Language 领域最新研究与进展

导语：在经典的 vision-language 任务上，能够增长的空间已经很小，已经过了暴力的通过数据去学习的阶段。真正的挑战其实是一些细分的领域。 雷锋网 AI 科技评论按：本文作者为阿德莱德大学助理教授吴琦，去年，他在为 AI 科技评论投递的独家稿件中回顾了他 ...
腾讯网

自动驾驶新风向：VLA（Vision-Language-Action）模型是如何崛起的？

2025年，随着智能驾驶开始往深度和广度两个方向去卷，智能驾驶行业往迎来一个显著信号：端到端大模型迈向2.0时代，VLA（Vision-Language-Action，视觉-语言-动作模型）或将成为国内车企全面竞争的焦点。 作为继VLM（视觉-语言模型）之后的进化形态，VLA通过整合 ...

Tensor Unveils OpenTau ( ) at CES 2026, Introducing a Breakthrough Open-Source Training ...

LAS VEGAS, Jan. 8, 2026 /PRNewswire/ -- At CES 2026, Tensor today announced the official open-source release of OpenTau ( ), a powerful AI training toolchain designed to accelerate the development of ...

Language shapes visual processing in both human brains and AI models, study finds

Neuroscientists have been trying to understand how the brain processes visual information for over a century. The development ...
news.ucsb.edu

The Intersection of Vision and Language

Nine thousand two hundred artificial intelligence researchers. Five thousand one hundred sixty-five research papers submitted, of which only 1,300 were accepted. One Best Student Paper. “Xin started ...
Slator

Cohere Labs Launches Vision-Language Dataset for African Languages

Cohere Labs unveils AfriAya, a vision-language dataset aimed at improving how AI models understand African languages and ...