Teaching Tiny VLA Models Where to Look and How to Move
On the map: Vision-language-action models → Generalist VLA models and benchmarks
Abstract (arXiv)
How can richer training supervision improve robot control while keeping the deployed policy compact? We present XS-VLA, a staged training framework using teacher-derived spatial labels and demonstration-conditioned action learning. Coarse-Grained Spatial Distillation (CSD) initializes the backbone through an auxiliary region-label task. Latent Flow Matching (LFM) then conditions an action-space velocity field on a demonstration latent, using KL regularization while jointly optimizing the backbone and action modules. The deployed policy contains 243.99M parameters and operates without the teacher or posterior encoder. XS-VLA achieves 90.25% average LIBERO success in each of two training seeds, compared with 86.00% for a SmolVLA-256M base trained under our settings. Ablations examine both training stages through matched image pretraining and Huber/MSE controls. On three Mobile ALOHA tasks, average strict success increases from 21.7% to 65.0%. These results demonstrate the control utility of auxiliary representation initialization and regularized demonstration-conditioned flow learning for compact VLA~policies.
Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Generalist VLA models and benchmarks, one of 13 directions in Vision-language-action models. Builds on 18 of its 22 references that are in the map. First posted on arXiv 5 Jul 2026.
Builds on
18 of 22 references.
Closest ancestors · 4
Same direction, newest first
- 2025SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics2 Jun
- 2024π0: A Vision-Language-Action Flow Model for General Robot Control31 Oct
- 2024OpenVLA: An Open-Source Vision-Language-Action Model13 Jun
- 2023RT-2: Vision-Language-Action Models Transfer Web Knowledge to Robotic Control28 Jul
Other references · 14
- 2022Flow Matching for Generative Modelingcited by 5,426
- 2023Diffusion Policy: Visuomotor Policy Learning via Action Diffusioncited by 2,730
- 2023Learning Fine-Grained Bimanual Manipulation with Low-Cost Hardwarecited by 1,490
- 2024Octo: An Open-Source Generalist Robot Policycited by 1,427
- 2023LIBERO: Benchmarking Knowledge Transfer for Lifelong Robot Learningcited by 1,338
- 2024SpatialVLM: Endowing Vision-Language Models with Spatial Reasoning Capabilitiescited by 679
- 2025SpatialVLA: Exploring Spatial Representations for Visual-Language-Action Modelcited by 502
- 2024Mobile ALOHA: Learning Bimanual Mobile Manipulation with Low-Cost Whole-Body Teleoperationcited by 377
- 2024TraceVLA: Visual Trace Prompting Enhances Spatial-Temporal Awareness for Generalist Robotic Policiescited by 298
- 2025SmolVLM: Redefining small and efficient multimodal modelscited by 226
- 2024TinyLLaVA: A Framework of Small-scale Large Multimodal Modelscited by 164
- 2025ThinkAct: Vision-Language-Action Reasoning via Reinforced Visual Latent Planningcited by 152
Built on by
Not cited yet.
Similar, not cited
Close in wording, no citation either way
- 2024TinyVLA: Towards Fast, Data-Efficient Vision-Language-Action Models for Robotic Manipulation
- 2026Offline Semantic Guidance for Efficient Vision-Language-Action Policy Distillation
- 2026FastOPD: On-Policy Distillation for Lightweight VLA Deployment
- 2025E0: Enhancing Generalization and Fine-Grained Control in VLA Models via Tweedie Discrete Diffusion
- 2026ActionGround: Training-Free Runtime Refinement of Frozen VLA Policies
- 2025Towards Deploying VLA without Fine-Tuning: Plug-and-Play Inference-Time VLA Policy Steering via Embodied Evolutionary Diffusion
- 2026AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Models
- 2025Goal-VLA: Image-Generative VLMs as Object-Centric World Models Empowering Zero-shot Robot Manipulation