@jotarunJotarun

RT @wenlong_huang: What representation enables open-world robot manipulation from generated videos? Introducing Dream2Flow, our recent work that bridges video generation and robot control with 3D object flow. https://t.co/vpRxoNBVF3 @Stanford #ICRA2026 1/N

RT @humanoidsdaily: 🚨 BREAKING: Amazon has officially acquired New York-based Fauna Robotics, the startup behind the soft-bodied Sprout humanoid. While Amazon has aggressively pursued automation in its fulfillment centers, this acquisition signals a massive escalation in the race for the consumer home market. Sprout is a bipedal robot designed specifically for the "messy reality" of shared human spaces.

RT @patrickhyin: We’re releasing OmniReset, a framework for training robot policies using large-scale RL and diverse resets for contact-rich, dexterous manipulation. OmniReset pushes the frontier of robustness and dexterity, without any reward engineering or demonstrations. Try the policies yourself in our interactive simulator! https://t.co/3hW3nYx2vD (1/N 🧵)

RT @StanfordMSL: π, But Make It Fly ✈️ We fine-tuned π0, a VLA model pretrained entirely on manipulators, to fly a drone that picks up objects, navigates through gates, and composes both skills from language commands.

RT @junfanzhu98: 📖Robotics World Model Reading Club #01 Summary @BostonDynamics, @Stanford, @AGIBOTofficial, @intbotai, @BytedanceTalk, @Google, @moonlake, @Rivian, @Meta, @Samsung, @UCBerkeley, @Cruise, @encord_team, @ManycoreTech, @OpenGraph_Labs, @neuralmotion, @AMD, @nvidia, @oysterecosystem, @Zoom, @FusionFundVC, @BoostVC, @yzilabs... policy learning→WM VLA: observation→action WAM: latent world→future trajectory→controllable action →Shift=reactive mapping→controllable simulation @nvidia Gr00t (7B, high mem efficiency on Thor)≈DreamDojo-style WAM. Bottleneck is NOT scale, but missing unified interface across perception–geometry–physics–action. 🧠 Representation Pixel space is redundant & non-geometric. Trend→Explicit 3D backbone: point cloud/mesh object+sub-object representations geometry-aware tracking (contact, affordance) Point-flow pipeline: detect→sample keypoint→track→dynamic graph Core tradeoff=which points&density (motion saliency/affordance attn) 🌍 4D Reconstructi→Unified Latent @GoogleDeepMind D4RT encodes video→temporally consistent latent field: geometry+motion+visibility unified Outputs: point clouds, 3D tracks, full reconstruct (300× faster) ❗Gap: no shared latent across: vision/geometry/semantics/action/physics ⚙️ Physics Gap Sim2Real Gap=physics, not vision: discontinuous contact deformable objects (∞ DoF) non-differentiable friction Engineering fails: brittle collision meshes, unstable contact Solutions: learned physics proxy hybrid pipeline convex decomposition (geometry → collision proxy, ~5× speedup) 🎥 Video Pretrain≠Interaction Video=strong prior but no counterfactuals Missing: force, depth, tactile, proprioception →can't answer: what if act differently ⏱️ Control≠Inference Real world=high-freq loop action chunking latent action FastWAM (train with rollout, infer without) KV-cache (AutoGaze) 👉control selects feasible trajectory, not full future modeling Thor is good, but LLM scaling≠robotics scaling 📉 Data No “robotics internet”: sim/video/teleop/factory logs fragmented no unified labeling or metrics Reality: factories use fixed primitives generalization often unnecessary Bitter lesson: data flywheel>pipelines (but robotics lacks one) 🦾 Embodiment Gap manipulation→full-body intelligence loco-manipulation+gaze+coordination Need cross-embodiment align (space, action, kinematics) 🔁 Sim2Real Pipeline human data→semantics→geometry→collision proxy→sim→fine-tuning Unsolved: deformables, contact stability, long horizon 🧩 Paper VQVAE (discrete latent) VL-JEPA (predictive align) token pruning (efficiency) recursive models (depth reuse) multi-path exploration (GRPO) ⚡ Infra→SLM Real-time stack (LLM infra too slow) →WM must compress into SLMs Future=small, domain-specialized, grounded models 🧪Bottlenecks no unified representation no data flywheel inference–control mismatch physics fragmented embodiment Reality can't be scraped like internet. It must be sensed, interacted, simulated. 👉 Goal: jointly optimize representation+simulation+action under physics constraints 💡minimal sufficient representation? can video DiT become WAM? vertical SLM inevitable? robotics ImageNet moment?

This basically shows how far we still are from a fully autonomous one.

RT @demishassabis: Excited to launch Gemma 4: the best open models in the world for their respective sizes. Available in 4 sizes that can be fine-tuned for your specific task: 31B dense for great raw performance, 26B MoE for low latency, and effective 2B & 4B for edge device use - happy building!

Most people don’t know how hard this is

RT @neuralink: We are working to restore mobility that was lost due to disease or spinal cord injury by allowing participants to control robotic arms with their thoughts. See how this is possible.

Someday

RT @awesomericky99: We present "hybrid system" that supplements conventional automation with "learning" for task & safety-level adaptiveness Deployed in factory for motor cable soldering (< 0.6 mm tolerance), resulting 108 motors, 99.4% SR with < 20 min data per task Paper: https://arxiv.org/abs/2604.22235

RT @BostonDynamics: Balancing commercial goals and robotics research can be tricky, but with Atlas we're making it work.

I am

RT @BostonDynamics: Everyone asks if Atlas can bring them a drink, but this robot can bring you the whole fridge. Using AI-driven behaviors, Atlas is doing hard work and coordinating its whole body to manage heavy objects, balancing complex contact points with accuracy and reliability.

RT @sama: OpenAI Robotics is hiring, looking for exceptional full-stack hardware, ops, systems, and ML engineers to help us program and manufacture robots that are useful for society. AI should be able to help people in the physical world. In the short term, we are focused on robots to support skilled workers to build our future infrastructure; in the long term, we imagine everyone having a personal robot doing anything they need. Our world simulation research program, led by Aditya Ramesh (@model_mechanic), has evolved over the past year into OpenAI Robotics. Progress is rapid, and based on a foundation of co-design between robotics hardware and ML research. If you love working hands-on across the robotics stack and want to build the future, please consider joining us. Send an email with your background and evidence of exceptional accomplishment to: robotics-recruiting@openai.com

RT @_tonytao_: What if some parts of a robot demonstration are more important than others? Most of a trajectory is free-space motion. But success or failure is often determined by a few critical moments around contact. In FACTR 2, we use force to find these moments and prioritize them for training. We find this helps policies learn better alignment and recovery behaviors, like the example below. w/ @StevenOh_ @JasonJZLiu 🧵(1/N)

Happy to see many of these robots don’t have legs.

RT @ruihangzhang: Introducing 📢📢ProxyPose📢📢 ✨Track 6-DoF motion of any query pixel via video-to-video translation 📰ArXiv: https://arxiv.org/abs/2607.06555 💻Code: https://github.com/ruihangzhang97/proxypose 🌐Webpage: https://ruihangzhang97.github.io/proxypose

RT @jeffacce: Your policy doesn't need 7B params. It simply needs dense features. Introducing Patch Policy: pretrained ViT + small transformer beats OpenVLA-OFT with 0.7% of its params, and trains on a 5090. Here it inserts a cable (~2mm tol), and does it again as we unplug mid-rollout. 🧵