Research map / Vision-language and multimodal / Robot action and navigation
Time Optimal Execution of Action Chunk Policies Beyond Demonstration Speed
Sunwoo Kim, Jeongjun Kim, Joseph J Lim
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| add time-optimal path parameterization (TOPP) (ablation) | RACE · Robomimic Square | speedup over demonstration: 1.42 -> 1.74 | Improved |
| add time-optimal path parameterization (TOPP) (ablation) | RACE · Robomimic Tool Hang | speedup over demonstration: 1.36 -> 1.86 | Improved |
| add time-optimal path parameterization (TOPP) (ablation) | RACE · Robomimic Square | success rate: 0.85 -> 0.86 | No clear effect |
| add test-time search (TTS) (ablation) | RACE · Robomimic Square | success rate: 0.81 -> 0.86 | Improved |
| add test-time search (TTS) (ablation) | RACE · Robomimic Tool Hang | success rate: 0.59 -> 0.62 | Improved |
| add test-time search (TTS) (ablation) | RACE · Robomimic Square | speedup over demonstration: 1.74 -> 1.74 | No clear effect |
| use desired states as actions without TOPP (State Fast-forward) | Action Fast-forward · Robomimic, no inference delay | speedup/success Pareto | No clear effect |
| replace Action Fast-forward (Async) with RACE under inference delay | Action Fast-forward (Async) · Robomimic Square and Tool Hang with 0.1 s delay | success rate vs speed | Improved |
| add inpainting to async fast-forward | Action Fast-forward (Async) · Robomimic with inference delay | success rate | No clear effect |
| replace SAIL with RACE | SAIL · Robomimic Tool Hang | success rate: 0.610 -> 0.715 | Improved |
| replace SAIL with RACE | SAIL · Robomimic Lift | speedup over demonstration: 2.520 -> 2.068 | Hurt |
| replace SAIL with RACE | SAIL · Robomimic Tool Hang | speedup over demonstration: 0.940 -> 2.053 | Improved |
Closest papers by meaning
- PACE: Phase-Aware Chunk Execution for Robot Policies with Action Chunking (NeurIPS 2026 · similarity 0.91)
- Real-Time Execution of Action Chunking Flow Policies (NeurIPS 2025 · similarity 0.89)
- FASTER: Rethinking Real-Time Flow VLAs (NeurIPS 2026 · similarity 0.89)
- Bidirectional Decoding: Improving Action Chunking via Guided Test-Time Sampling (ICLR 2025 · similarity 0.89)
- SkiP: When to Skip and When to Refine for Efficient Robot Manipulation (NeurIPS 2026 · similarity 0.89)
- ElegantVLA: Learning When to Think for Efficient Vision-Language-Action Models (NeurIPS 2026 · similarity 0.87)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
