Search

5 results for “LIBERO”

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: 7.78 s Astra request latency (median) (as reported)

   3 Nov 2026Huawei Inc.

2. 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: 75.6% average success (as reported)

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

3. Research · Generative media

   One-step action generation and asynchronous pipeline achieve real-time world action models

   RealtimeWAM: One-Step Asynchronous World Action Models

   RealtimeWAM uses teacher-anchored consistency distillation and cross-expert pipelining to accelerate Mixture-of-Transformers based world action models by up to 25x on H100 GPUs with less than 1% accuracy loss

   LIBERO: 97% success rate (as reported)

   6 Oct 2026Nanyang Technological University, Beihang University, Sensetime

4. Research · Robotics & embodied AI

   Agent-side memory steers stateless robots through long manipulation tasks

   2AM: Grounding Agent-Side Memory as Guidance for Steerable Action Models in Long-Horizon Manipulation

   2AM separates task memory from motor control, using language and optional 2D hints to guide a single RGB-only action model.

   LIBERO-Mem: 76.3% average completion (as reported)

   12 Sept 2026KU Leuven, Meituan Inc., Flanders Make@KU Leuven

5. Research · Robotics & embodied AI

   Recurrent memory policy improves robot manipulation on tasks requiring spatial recall, counting, and timing

   ReCAT: Remember, Count, and Time: Structured Recurrent Memory for Robot Manipulation

   ReCAT combines structured recurrent memory with direct current observation to achieve high success on benchmark and real-world tasks

   LIBERO: 95.3% success rate (as reported)

   5 Oct 2026University of Edinburgh, Karlsruhe Institute of Technology, NVIDIA