4arXiv cs.AI (Artificial Intelligence)·
JoyNexus: Multi-tenant post-training service for Vision-Language-Action models
JoyNexus is a proposed unified service architecture for multi-tenant supervised fine-tuning, reinforcement learning, and evaluation of Vision-Language-Action (VLA) models. The system decouples training, inference, and environment services behind APIs, with shared base models and tenant-isolated policy slots, scheduled via global training and inference queues. A group-batching technique enables a single shared backbone forward pass over heterogeneous VLA data from multiple tenants, reducing aggregate GPU time compared to isolated single-tenant execution. The work targets the inefficiency of exclusive GPU allocation for short or bursty robotics workloads.
Related guides (2)
Related events (8)
LabVLA: Vision-Language-Action model and RoboGenesis data engine for scientific laboratory robotics
Researchers introduce LabVLA, a Vision-Language-Action model designed to bridge written scientific protocols and physical robot execution in laboratory settings. To address the data scarcity problem, they build RoboGenesis, a simulation-based data engine that composes lab workflows from atomic skills and generates structured demonstrations across robot embodiments. LabVLA uses a two-stage training recipe combining FAST action token pretraining on a Qwen3-VL-4B-Instruct backbone with flow matching posttraining via a DiT action expert. On the LabUtopia benchmark, LabVLA achieves the highest average success rate among evaluated baselines in both in-distribution and out-of-distribution settings.
Bringing Robotics AI to Embedded Platforms: Dataset Recording, VLA Fine-Tuning, and On-Device Optimizations
NXP and Hugging Face describe a pipeline for deploying Vision-Language-Action (VLA) models on embedded/edge hardware, covering dataset recording, fine-tuning, and on-device optimization techniques. The post targets robotics applications where inference must run on resource-constrained microcontrollers or SoCs rather than cloud GPUs. Key topics include quantization, model compression, and integration with the LeRobot ecosystem. This represents a practical engineering bridge between frontier VLA research and real-world embedded robotics deployment.
Qwen-VLA: Unified Vision-Language-Action Model Across Robot Tasks, Environments, and Embodiments
Alibaba's Qwen team presents Qwen-VLA, a unified embodied foundation model that extends the Qwen vision-language stack to continuous action and trajectory generation via a DiT-based action decoder. The model is jointly pretrained on diverse data spanning manipulation trajectories, egocentric demonstrations, synthetic simulation, and navigation data, with embodiment-aware prompt conditioning to support multiple robot platforms. A unified action-and-trajectory prediction framework covers manipulation, navigation, and trajectory prediction tasks. Benchmarks show strong results: 97.9% on LIBERO, 73.7% on Simpler-WidowX, 69.0% OSR on R2R navigation, and 76.9% average OOD success in real-world ALOHA experiments.
CHORUS: Single VLA policy enables decentralized multi-robot collaboration without inter-robot communication
CHORUS is a framework that adapts a single vision-language-action (VLA) backbone to control diverse multi-robot teams in a fully decentralized manner, with each robot running an independent copy conditioned only on its own observations and a robot-identifying prompt. Real-world experiments across tasks like tape measurement, book handovers, and laundry basket lifting show a 64-percentage-point improvement over decentralized from-scratch models and 40-point improvement in reactivity to teammate behavior, while outperforming centralized baselines. The key insight is that pretrained VLA visuomotor priors are sufficient to enable reactive coordination without explicit inter-robot communication or alignment procedures at inference time.
Act2Answer: Benchmarking commonsense and world knowledge retention in Vision-Language-Action models
Researchers introduce Act2Answer, a protocol for evaluating how much commonsense and factual knowledge VLA models retain after fine-tuning on robotics data. The approach converts knowledge benchmark questions into tabletop object-placement episodes, yielding action-grounded success rates that reduce confounds from low-level control failures. A large-scale study of 7 VLA models and 9 VLM baselines finds that VLAs retain solid performance on simple concepts but show larger gaps on richer semantic categories compared to their source VLMs, and that VQA co-training is associated with better knowledge retention.
Task-Agnostic Pretraining (TAP) decouples motor learning from language grounding in VLA models
Researchers propose Task-Agnostic Pretraining (TAP), a two-stage framework for Vision-Language-Action models that separates physical motor skill acquisition from semantic language alignment. The first stage learns motor priors from cheap unlabeled interaction data via a self-supervised Inverse Dynamics objective; the second stage grounds these priors in language using minimal expert demonstrations. On the SIMPLER benchmark, TAP matches models trained on over 1M expert trajectories while using orders of magnitude less labeled data, and on a real-world WidowX robot retains 25% success under camera perturbations where internet-scale baselines collapse to 0%.
FORCE: Efficient RL fine-tuning for Vision-Language-Action models via value-calibrated warm-up and self-distillation
Researchers introduce FORCE, a 3-stage reinforcement learning fine-tuning framework for Vision-Language-Action (VLA) models that addresses sample inefficiency caused by unstable Q-functions and low-quality exploration data. The framework uses a Value-Calibrated Warm-Up phase followed by Q-function-filtered policy updates, eliminating the need for costly human interventions during training. Evaluated on simulation and real-world robotic tasks, FORCE achieves a 79% absolute improvement in task success rates, outperforms prior RL methods by 10%, and accelerates training by 32.5%.
HABC: Hierarchical Advantage Weighting for Online RL Fine-Tuning of Vision-Language-Action Policies
Researchers introduce Hierarchical Advantage-Weighted Behavior Cloning (HABC), a method for fine-tuning pretrained Vision-Language-Action (VLA) policies via online RL using only sparse binary episode outcomes. HABC trains separate critic heads for viability and efficiency objectives, combines them via a state-adaptive gate, and applies intervention-aware credit assignment to avoid incorrect supervision across human-intervention boundaries. On three contact-rich bimanual real-robot tasks, HABC improves success rates from SFT baselines of 36%, 44%, and 12% to 92%, 88%, and 38% respectively. The work addresses a fundamental credit assignment problem in robot learning from sparse outcome signals.