Research map / Image generation / Autoregressive and latent generation
MSP: Probabilistically Consistent Multi-Scale Action Generation
Zhixuan Lin, Gengqi Liu, Chao Zheng, Gao Lin, Jindong Yu, Song Gao, Fei Wang
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 |
|---|---|---|---|
| keep all intermediate scales [1,2,4,8] instead of dropping scale 2 (ablation) | MSP · LIBERO-Long | success rate (%): 91.6 -> 96.8 | Improved |
| use dense scales instead of sparse scales [1,4,16] with long action length (ablation) | MSP · LIBERO-Long | success rate (%): 79.4 -> 90.2 | Improved |
| use downsampling factor 4 instead of 2 (ablation) | MSP · LIBERO-Long | success rate (%): 89.6 -> 96.8 | Improved |
| use downsampling factor 4 instead of 8 (ablation) | MSP · LIBERO-Long | success rate (%): 93.0 -> 96.8 | Improved |
| use action length 32 instead of 16 (ablation) | MSP · LIBERO-Long | success rate (%): 90.8 -> 96.8 | Improved |
| use action length 32 instead of 64 with extra scale (ablation) | MSP · LIBERO-Long | success rate (%): 90.2 -> 96.8 | Improved |
| use MeanFlow instead of diffusion or standard flow matching for scale-wise generation (ablation) | MSP · LIBERO-Long | success rate (%) | Improved |
| map actions to continuous VAE latent space rather than raw action space (ablation) | MSP · LIBERO-Long | success rate (%) | Improved |
| add scale-wise autoregressive modeling (ablation) | MSP · LIBERO-Long | success rate (%) | Improved |
| use MSP instead of QueST for few-shot adaptation | MSP · LIBERO-Long few-shot (5 demos) | success rate (%): 66.7 -> 76.4 | Improved |
| use MSP in place of DSP on LIBERO-Goal | MSP · LIBERO-Goal | success rate (%): 90.8 -> 93.8 | No clear effect |
| use MSP instead of DSP in terms of parameters | MSP · LIBERO-Long | action-head parameters: about 60% of DSP parameters | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- CARP — CARP: Compression Through Adaptive Recursive Partitioning for Multi-Dimensional Images (CVPR 2020) · better on 1 of 1 shared result
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Chain-of-Action: Trajectory Autoregressive Modeling for Robotic Manipulation (NeurIPS 2025 · similarity 0.88)
- Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry (NeurIPS 2026 · similarity 0.88)
- Latent Diffusion Planning for Imitation Learning (ICML 2025 · similarity 0.88)
- Quantization-Free Autoregressive Action Transformer (NeurIPS 2025 · similarity 0.88)
- FreqPolicy: Efficient Flow-based Visuomotor Policy via Frequency Consistency (NeurIPS 2025 · similarity 0.88)
- OSVI-WM: One-Shot Visual Imitation for Unseen Tasks using World-Model-Guided Trajectory Generation (NeurIPS 2025 · similarity 0.88)
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.
