LaST-R1
PulseAugur coverage of LaST-R1 — every cluster mentioning LaST-R1 across labs, papers, and developer communities, ranked by signal.
- 2026-05-11 research_milestone LaST-R1 AI model achieves 99.9% success on the LIBERO benchmark and outperforms previous SOTA models by 22.5% in real-world tasks. 来源
1 天有情绪数据
-
LaST-R1 AI achieves 99.9% success on LIBERO benchmark
Researchers have developed a novel embodied AI training method that integrates latent space physical reasoning. This new paradigm, named LaST-R1, has demonstrated exceptional performance, achieving 99.9% success on the …
-
新的 VLA 模型 LaST-R1 和 DIAL 通过高级推理增强机器人操作
两篇新的研究论文介绍了用于机器人操作的先进视觉-语言-动作 (VLA) 模型。LaST-R1 将潜在的思维链推理与强化学习相结合，以提高适应性和泛化能力，在 LIBERO 基准测试中取得了 99.8% 的成功率。DIAL 通过潜在世界模型将高级意图与低级动作执行解耦，使其能够以少 10 倍的演示进行学习，并泛化到现实世界任务。
-
New research boosts LLM reasoning with speculative methods and physical insights
Recent research explores novel methods to enhance the reasoning capabilities and efficiency of large language models (LLMs). Papers introduce techniques like speculative exploration for Tree-of-Thought reasoning to brea…