Search

7 results for “LIBERO-PRO”

1. Research · Efficiency & systems

   Robot fleets can finish more work with fewer active robots

   Serving Agentic Robot Fleets: Workload Characterization, KV-Cache Thrashing, and Admission Control

   In sustained replay of robot workloads, limiting active episodes restored shared computing throughput and roughly quadrupled completed programs.

   LIBERO-PRO: 0.81 median turn rate per second of environment time (as reported)

   3 Nov 2026Huawei Inc.

2. Research · Robotics & embodied AI

   Synthetic demonstrations let robot policies escape sparse-reward failures

   SynthDemo-RL: Breaking the Zero-Reward Barrier in VLA Adaptation with LLM-Guided Synthetic Demonstrations

   An automated teacher supplied successful manipulation trajectories that enabled subsequent reinforcement learning on tasks where direct PPO received no reward.

   LIBERO-PRO: 97.1% success rate (as reported)

   21 Sept 2026Fujitsu Limited, The Institute of Statistical Mathematics

3. Research · Robotics & embodied AI

   Better video predictions improve robots’ performance on unfamiliar manipulation tasks

   Video Prediction Policy 2: Predict Better, Act Better

   VPP2 combines manipulation-focused video pretraining with action learning while preserving the video model’s ability to generalize.

   LIBERO-Pro: 45% overall success rate (as reported)

   9 Oct 2026Robotera, Tsinghua University, HKUST (GZ)

4. Research · Robotics & embodied AI

   Visual action rehearsal helps language models control robot manipulation

   World Action Agent: Harnessing VLMs for Robot Manipulation via World Action Rehearsal

   A multi-agent harness lets vision-language models preview, revise, and visually correct robot actions before and during execution.

   LIBERO-Pro: 31.05 calls average model calls per episode (as reported)

   25 Sept 2026HKUST(GZ), CUHK, Knowin AI

5. Research · Code & software

   One demonstration can bootstrap reusable robot manipulation programs

   RAPID: Robot Agentic Programming from Demonstrations

   RAPID infers task goals, action primitives and a verification environment from a visual demonstration, then iteratively refines executable robot code.

   27 Sept 2026Massachusetts Institute of Technology, National University of Singapore, University of Pennsylvania

6. Research · Robotics & embodied AI

   Adaptive action chunking improves robot control by varying horizon based on prediction reliability.

   GeoAAC: Geometry-Based Adaptive Action Chunking from Denoising Trajectories in VLA Policies

   GeoAAC uses the geometry of flow matching denoising trajectories to adjust the action chunk size per step, boosting real-world success rates from 53.3% to 74.4%.

   real-world manipulation tasks: 74.4% success rate (as reported)

   20 Sept 2026Tongji University, The Hong Kong Polytechnic University, University of Electronic Science and Technology of China

7. Research digest

   Agent Safety Moves From Prompts to Process-Level Controls and Tests

   The week’s clearest result is that reliable agents require runtime constraints, process-aware evaluation, and targeted correction—not stronger instructions alone.

   24 Sept 2026