SALA: Semantically Aligned Latent Actions Enable Dual-Axis Scaling of Embodied Foundation Models
cs.RO, cs.AI, cs.LG
Submitted: 2026-11-12
Updated: 2026-11-15
Project page: https://sala-vla.github.io
Terminology
Sources
- Qwen3-VL Technical Report
- Video PreTraining (VPT): Learning to Act by Watching Unlabeled Online Videos
- Latent Action Diffusion for Cross-Embodiment Manipulation
- $\pi_0$: A Vision-Language-Action Flow Model for General Robot Control
- RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control
- Genie: Generative Interactive Environments
- UniVLA: Learning to Act Anywhere with Task-centric Latent Actions
- UniT: Toward a Unified Physical Language for Human-to-Humanoid Policy Learning and World Modeling
- RoboTwin 2.0: A Scalable Data Generator and Benchmark with Strong Domain Randomization for Robust Bimanual Robotic Manipulation
- IGOR: Image-GOal Representations are the Atomic Control Units for Foundation Models in Embodied AI
- Moto: Latent Motion Token as the Bridging Language for Learning Robot Manipulation from Videos
- ConLA: Contrastive Latent Action Learning from Human Videos for Robotic Manipulation
- XR-1: Towards Versatile Vision-Language-Action Models via Learning Unified Vision-Motion Representations
- METIS: Multi-Source Egocentric Training for Integrated Dexterous Vision-Language-Action Model
- EgoMimic: Scaling Imitation Learning via Egocentric Video
- Emergence of Human to Robot Transfer in Vision-Language-Action Models
- OpenVLA: An Open-Source Vision-Language-Action Model
- Learning to Act from Actionless Videos through Dense Correspondences
- EgoWAM: World Action Models Beyond Pixels with In-the-Wild Egocentric Human Data
- CLAM: Continuous Latent Action Models for Robot Learning from Unlabeled Demonstrations
Related papers
- Rethinking Actions: Continuity-Aware Priors for World-Model Planning
- MOSAIC: Modular Obstacle and Spatial-Aware Framework for Manipulation In Cluttered Scenes
- SAGAS: Semantic-Aware Graph-Assisted Stitching for Offline Temporal Logic Planning
- DexSeed: Expanding Human Hand-Object Interactions into Robot Demonstrations for Dexterous Policy Learning
- LaST0.5: A Generalist World Action Model with Interaction-Centric Latent Reasoning
- CIALLO: Learning Shared Intent Representations for Cross-Embodiment Manipulation