ESPADA: Execution Speedup via Semantics Aware Demonstration Data Downsampling for Imitation Learning

arXiv:2512.07371v3 Announce Type: replace-cross Abstract: Behavior-cloning based visuomotor policies enable precise manipulation but often inherit the slow, cautious tempo of human demonstrations, limiting practical deployment. However, prior studies on acceleration methods mainly rely on statistical or heuristic cues that ignore task semantics and can fail across diverse manipulation settings. We present ESPADA, a semantic and spatially aware framework that segments demonstrations using a VLM-LLM pipeline with 3D gripper-object relations, enabling aggressive downsampling only in non-critical segments while preserving precision-critical phases, without requiring extra data or architectural modifications, or any form of retraining. To scale from a single annotated episode to the full dataset, ESPADA propagates segment labels via Dynamic Time Warping (DTW) on dynamics-only features. Across both simulation and real-world experiments with ACT and DP baselines, ESPADA achieves approximately a 2x speed-up while maintaining success rates, narrowing the gap between human demonstrations and efficient robot control.

Related

• Learning Multi-Modal Whole-Body Control for Real-World Humanoid Robots
• From Seeing to Simulating: Generative High-Fidelity Simulation with Digital Cousins for Generalizable Robot Learning and Evaluation
• ExpertGen: Scalable Sim-to-Real Expert Policy Learning from Imperfect Behavior Priors
• No More Marching: Learning Humanoid Locomotion for Short-Range SE(2) Targets

Source: arXiv cs.AI | 2026-04-28