模型官方一手
Pathway在Amazon SageMaker HyperPod开发大脑架构
Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod
精选理由
Pathway在AWS平台上开发BDH模型，在ARC-AGI-1基准上创下成本效率新高。
Pathway的Baby Dragon Hatchling (BDH)是一种受大脑启发的后Transformer架构，在潜在空间中推理而非输出思维链标记。Pathway在Amazon SageMaker HyperPod上开发和扩展了BDH模型。BDH-CQ在ARC-AGI-1基准测试上创造了新的成本效率记录。
原文 · AWS Machine Learning Blog
Pathway’s brain-inspired architecture development on Amazon SageMaker HyperPod
Pathway's Baby Dragon Hatchling (BDH) is a brain-inspired, post-transformer architecture that reasons in latent space instead of emitting chain-of-thought tokens. See how Pathway develops and scales BDH on Amazon SageMaker HyperPod, and how BDH-CQ set a new cost-efficiency mark on the ARC-AGI-1 benchmark.
