# Learning Semantic Atomic Skills for Multi-Task Robotic Manipulation

arXiv: https://arxiv.org/abs/2512.18368 · HTML: https://mlsanity.com/paper/2512.18368/

## Key facts

- Cited by 5 papers in the map (data through 14 Oct 2026).
- Placed in Robot manipulation imitation learning, one of 13 directions in Offline RL and imitation learning from data.
- First posted on arXiv 20 Dec 2025.

## Placed in

- Offline RL and imitation learning from data / [Robot manipulation imitation learning](https://mlsanity.com/offlinerl/robot-manipulation-imitation-learning/)

## Built on by (most cited first, 5 of 5)

- [AC-VLA: Robust Out-of-Distribution Action Execution via Compositional Learning](https://mlsanity.com/paper/2607.15714/) (2026)
- [Improving Vision-Language-Action Model Fine-Tuning with Structured Stage and Keyframe Supervision](https://mlsanity.com/paper/2606.26801/) (2026)
- [Teach and Grow: An Agent-Centered Architecture for General Robot Learning](https://mlsanity.com/paper/2608.17209/) (2026)
- [REFACTOR-VLA: Unsupervised Library Learning of Typed Motor Programs](https://mlsanity.com/paper/2609.01215/) (2026)
- [OGPO: Offline Goal-conditioned Policy Optimization for Recoverable Vision-Language-Action Models](https://mlsanity.com/paper/2610.12914/) (2026)

## Abstract (arXiv)

Scaling imitation learning to diverse multi-task robot manipulation remains challenging due to suboptimal demonstrations, behavioral multi-modality, and destructive interference across tasks. While skill-based methods offer a promising direction by decomposing behaviors into reusable abstractions, existing approaches often learn skills that are either biased toward linguistic structure or lack semantic alignment across tasks, limiting generalization. In this work, we propose AtomSkill, a novel framework that learns a semantically aligned Atomic Skill Space from demonstrations and enables robust long-horizon execution through keypose imagination. Our method introduces: (1) semantic contrastive skill alignment, which partitions demonstrations into variable-length atomic skills and employs a contrastive objective to jointly enforce semantic consistency and temporal coherence, yielding a compact and reusable skill library; and (2) action decoding with keypose imagining, where the policy predicts both a skill's terminal keypose and immediate actions, thereby supporting progress-aware skill transitions. During inference, an atomic skill diffusion sampler generates plausible skill sequences, while predicted keyposes autonomously trigger smooth skill chaining. Extensive experiments in simulation and real-world settings show that AtomSkill consistently outperforms state-of-the-art imitation learning and skill-based baselines. Project page: https://atom-skill.github.io.