Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models 文章

ArXiv CS.CV2026-06-16NEWSen作者: Dianqiao Lei, Lianlei Shan
查看原文 →
关系图谱

Think Less, Act Early: Reinforced Latent Reasoning with Early Exit in Vision-Language-Action Models · 相关技术

相关技术

Vision-Language-Action modelReinforcement Learning-based Denoisinglatent reasoningChain-of-thought reasoningreinforcement learningchain-of-thought