Research map / Continual learning / Online and replay-based
Lifelong Language-Conditioned Robotic Manipulation Learning
Xudong Wang, Zebin Han, Zhiyu Liu, Gan Li, Jiahua Dong, Baichen Liu, Lianqing Liu, Zhi Han
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 |
|---|---|---|---|
| SkillsCrafter vs SD-LoRA + SkSA | SkillsCrafter · LLCRM | Average success rate (%): 50.0 -> 52.0 | Improved |
| SkillsCrafter vs SD-LoRA + SkSA, forgetting | SkillsCrafter · LLCRM | Average forgetting rate (%): 20.8 -> 16.0 | Improved |
| add Gumbel-softmax gating sparse LoRA loading (ablation) | SkillsCrafter · LLCRM | Average success rate (%): 46.9 -> 52.0 | Improved |
| add inheritance of A (shared knowledge) (ablation) | SkillsCrafter · LLCRM | Average success rate (%): 48.2 -> 52.0 | Improved |
| add skills orthogonal term on B (ablation) | SkillsCrafter · LLCRM | Forgetting rate (%): 22.9 -> 16.0 | Improved |
| use SVD subspace instead of average pooling of instruction encodings (ablation) | SkillsCrafter · LLCRM | Average success rate (%): 44.4 -> 52.0 | Improved |
| weighted aggregation of adapters instead of top-1 selection (ablation) | SkillsCrafter · LLCRM | Average success rate (%): 50.4 -> 52.0 | Improved |
| instruction matching instead of vision matching (ablation) | SkillsCrafter · LLCRM | Average success rate (%): 48.9 -> 52.0 | Improved |
Closest papers by meaning
- Think Small, Act Big: Primitive Prompt Learning for Lifelong Robot Manipulation (CVPR 2025 · similarity 0.90)
- Goal-Oriented Skill Abstraction for Offline Multi-Task Reinforcement Learning (ICML 2025 · similarity 0.88)
- Phoenix: A Motion-based Self-Reflection Framework for Fine-grained Robotic Action Correction (CVPR 2025 · similarity 0.88)
- SkillDiffuser: Interpretable Hierarchical Planning via Skill Abstractions in Diffusion-Based Task Execution (CVPR 2024 · similarity 0.88)
- STAR: Learning Diverse Robot Skill Abstractions through Rotation-Augmented Vector Quantization (ICML 2025 · similarity 0.87)
- Language Guided Skill Discovery (ICLR 2025 · 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.
