SmolVLA-MoE introduces a sparse Mixture-of-Experts (MoE) architecture into Vision‑Language‑Action (VLA) models for robotic manipulation, replacing dense feed‑forward networks with multiple expert FFNs and a token‑level router. The framework includes a shared expert with a tunable weight to balance pre‑trained knowledge and task‑specific specialization, activating only one expert during inference to keep the parameter count low. Evaluated on the LIBERO benchmark, SmolVLA‑MoE achieves an average success rate of 84.25%, surpassing both SmolVLA and OpenVLA‑7B while maintaining comparable inference efficiency.
arXiv:2606. 20246v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models pre-trained on massive video-robot datasets have revolutionized robotic manipulation, yet their multi-billion parameter architectures impose prohibitive computational burdens during downstream fine-tuning and real-time inference.
arXiv:2507. 05116v5 Announce Type: replace-cross Abstract: Recent large-scale Vision Language Action (VLA) models have shown superior performance in robotic manipulation tasks guided by natural language.
The paper introduces VLAct, a Vision‑Language‑Action model that focuses on representation‑centric continued pre‑training rather than merely scaling robot data. VLAct is trained on diverse, multi‑embodiment robot data and preserves a broad VLM prior while encouraging shared action semantics across embodiments. Experiments across simulation, real‑world, and unseen‑embodiment settings show that VLAct consistently outperforms existing industrial VLA systems, achieving high success rates with only a modest compute budget and open‑source data.
arXiv:2605.13119v2 Announce Type: replace-cross Abstract: Vision-language-action (VLA) models are effective robot action executors, but they remain limited on long-horizon tasks due to the dual burde...
arXiv:2610.08133v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models achieve strong robotic manipulation performance but incur high computational costs from processing long token seq...
StarVLA-α is a streamlined Vision‑Language‑Action (VLA) model that reduces architectural and pipeline complexity to facilitate systematic study of VLA design choices. By employing a strong VLM backbone and minimal design, it achieves competitive performance across multiple benchmarks (LIBERO, SimplerEnv, RoboTwin, RoboCasa) and outperforms the baseline π₀.₅ by 20% on the RoboChallenge benchmark. The authors plan to release the code to support future VLA research.
Vision-language-action (VLA) models commonly adopt an LLM-centric $V \to L \to A$ pathway, where visual observations are projected into the representation space of a large language model before being decoded into robot actions. Although effective, this design incurs substantial computation and memory overhead at every policy invocation.
MambaAct introduces a compact Mamba-based action expert to replace the transformer-based expert in flow‑matching vision‑language‑action models, preserving the pretrained visual‑language prefix encoder while reducing the action expert size by tenfold. The new design employs a lightweight Context Compressor to embed vision‑language conditioning into each Mamba block, eliminating repeated token‑level cross‑attention. Experiments show a 64.8 % reduction in denoising latency on LIBERO and 58.8 % in real‑world deployment, with a modest 1.08 % increase in mean success rate across multiple fine‑tuning seeds.
arXiv:2607. 04171v1 Announce Type: cross Abstract: Large Vision-Language Models (LVLMs) have shown strong multimodal understanding and spatial grounding, but their computational cost limits real-time robotic control.
arXiv:2608. 11739v1 Announce Type: cross Abstract: The prevailing recipe for Vision-Language-Action (VLA) models couples a pretrained VLM with a separately trained flow-matching action expert.
NestedVLA proposes a new approach for consolidating skills in Vision‑Language‑Action models by learning skills independently and then generating task‑specific parameters via a nested hypernetwork, rather than aggregating parameters. The method keeps a frozen base VLA and task‑specific compressed VLM residuals, enabling flexible, interference‑mitigated skill reuse. Experiments on MetaWorld, ManiSkill, and CALVIN show that NestedVLA outperforms the best model‑merging baseline by 41 pp on average and approaches per‑task fine‑tuning performance.
arXiv:2606. 08530v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models achieve strong benchmark performance but still struggle in real-world deployment with unseen objects, background shifts, and different robot embodiments.