Pathway的BDH-CQ用极低成本在ARC-AGI 1拿到29.5%，靠的是潜在空间推理而不是CoT，还验证了600B参数规模，值得一看。
Pathway的BDH-CQ模型在ARC-AGI 1基准上获得29.5%的分数，但每个任务成本仅为0.0007美元。该模型通过潜在空间中的循环推理而非链式思维（CoT）来实现这一成绩。研究团队还验证了类似Transformer的扩展规律，参数规模可达6000亿。这一成果展示了在成本效率上的显著优势。
This is kind of wild. BDH-CQ scored 29.5% on ARC-AGI 1. Fair. But with just $.0007 per task. How ...
This is kind of wild. BDH-CQ scored 29.5% on ARC-AGI 1. Fair. But with just $.0007 per task. How it’s done is particularly interesting. It reasons recurrently in latent space rather than using CoT. They’ve also verified Transformer-like scaling up to 600B params. Zuzanna Stamirowska @zuzanna_pathway Pathway’s BDH-CQ model redefines the Cost-Efficiency Frontier on ARC-AGI-1: $.0007 at 29.5%. This is made possible by in-context learning and latent reasoning. Welcome to the Post-Transformer Era. 🔗 View Quoted Tweet 💬 3 🔄 4 ❤️ 14 👀 1851 📊 5 ⚡