Research map / Reinforcement learning / Visual and representation RL
Masked Skill Token Training for Hierarchical Off-Dynamics Transfer
Zeyu Feng, Haiyan Yin, Yew-Soon Ong, Harold Soh
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 masked skill tokens and feasibility-aware planning (MSTT) versus Diffuser | MSTT · Maze2D (avg) | return: 33.12 -> 140.35 | Improved |
| fine-tune Diffuser on state-only target demonstrations | Diffuser · Maze2D (avg) | return: 33.12 -> 57.11 | Improved |
| MSTT versus state-only fine-tuned Diffuser on goal rate | MSTT · Maze2D (avg) | goal success rate: 72% -> 94.33% | Improved |
| MSTT versus action-labelled BC-ta on goal rate | MSTT · Maze2D (avg) | goal success rate: 82% -> 94.33% | Improved |
| MSTT versus fine-tuned Diffuser on Maze2D Target 1 goal rate | MSTT · Maze2D Target 1 | goal success rate: 99% -> 89% | Hurt |
| MSTT versus Diffuser on FetchReach goal rate | MSTT · FetchReach | goal success rate: 99% -> 88% | Hurt |
| MSTT versus Diffuser on FetchReach failure rate | MSTT · FetchReach | failure rate: 61% -> 0% | Improved |
| MSTT versus inverse dynamics on ReplicaCAD goal rate | MSTT · Habitat ReplicaCAD | goal success rate: 23% -> 75% | Improved |
| increase VQ-VAE codebook size from 15 to 240 (ablation) | MSTT · Maze2D Target 2 | goal success rate: 0% -> 95% | Improved |
| increase VQ-VAE codebook size from 240 to 1000 (ablation) | MSTT · Maze2D Target 3 | goal success rate: 99% -> 86% | Hurt |
| increase VQ-VAE codebook size from 240 to 512 (ablation) | MSTT · Maze2D Target 2 | goal success rate: 95% -> 90% | No clear effect |
| use VLM-generated demonstration instead of provided demonstration | MSTT · Maze T1 | goal success rate: 89% -> 94% | No clear effect |
Closest papers by meaning
- Robust Policy Learning via Offline Skill Diffusion (AAAI 2024 · similarity 0.89)
- Foundation Policies with Hilbert Representations (ICML 2024 · similarity 0.89)
- Diffusion Subgoal Planning for Long-Horizon Offline Goal-Conditioned Reinforcement Learning (NeurIPS 2026 · similarity 0.89)
- Learning from Reward-Free Offline Data: A Case for Planning with Latent Dynamics Models (NeurIPS 2025 · similarity 0.88)
- Flow Matching for Offline Reinforcement Learning with Discrete Actions (NeurIPS 2026 · similarity 0.88)
- Learning Versatile Skills with Curriculum Masking (NeurIPS 2024 · 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.
