Kyle🤖🚀🦭

Most capable generalist robotics models today are closed or at best, open weights. But robotics won’t reach its ChatGPT moment without real openness. That GPT moment was built on years of open tools and datasets such as Python, PyTorch, ImageNet and more, that let researchers inspect, reproduce, and build. Today, we’re introducing MolmoAct 2: a fully open-source action reasoning model for real-world robotics. We rethought and reshaped everything! 🧵👇

I feel it’s really unhelpful that searching for “deep RL” sends you to Q learning, MDPs, Bellman’s equation etc, when it’s literally just Run LLM agent on data -> was it good? -> policy gradient +/- reward Like that’s actually it! And LLMs are just stacks of attn+MLP

Watch Spot crouch, jump, climb boxes and leap across gaps, controlled by a neural network trained with reinforcement learning (RL) and multi-expert distillation. Multiple expert policies were trained and distilled together into a single policy that was fine tuned to improve performance over diverse terrains. This work was inspired by ANYmal’s parkour capabilities. The neural network processes depth data from Spot's sensors to construct an understanding of the environment.

My robot can now feel how hard it's gripping something. I didn't add any sensors. Comment "tactile" and I'll DM how it works.