arXiv AI
Jul 28
VLASH: Real-Time VLAs via Future-State-Aware Asynchronous Inference
arXiv:2512. 01031v2 Announce Type: replace-cross Abstract: Vision-Language-Action models (VLAs) are becoming increasingly capable across diverse robotic tasks.
DEFLECT: Temporal Counterfactual Preference Learning for Delay-Robust Asynchronous VLAs
arXiv:2605. 19294v2 Announce Type: replace-cross Abstract: Vision-Language-Action (VLA) policies increasingly rely on asynchronous inference to hide large-model latency behind ongoing robot motion.
Related stories
arXiv Machine Learning
Sep 17
Reinforcement Learning for Real-Time Vision-Language-Action Policies
The paper presents Real‑Time EXPO‑FT, a reinforcement learning framework that fine‑tunes large Vision‑Language‑Action models for real‑time robotic control. It separates slow, expressive action generation from fast, reactive edits, allowing a lightweight policy to adjust actions based on the latest observation. Experiments on the Kinetix benchmark and four dynamic real‑world tasks show that Real‑Time EXPO‑FT achieves superior performance, improving policy success rates from 42% to 97% with only ten minutes of online data and no human intervention.
arXiv Machine Learning
Aug 26
Learning to Act While Waiting: RL Finetuning of Generalist Robot Policies Under Inference Latency
The paper introduces ARLI, a latency‑aware framework that enables reinforcement learning fine‑tuning of large generalist robot policies despite inference delays. ARLI combines asynchronous inference with state augmentations—incorporating committed actions and mid‑inference observations—to restore near‑Markovian dynamics and maintain reactivity. Experiments on simulated and real‑world manipulation tasks show that ARLI allows effective policy improvement under latency, outperforming standard RL even in no‑latency scenarios.
arXiv AI
Aug 18
EvoScene-VLA: Evolving Scene Beliefs Inside the Action Decoder for Chunked Robot Control
arXiv:2605. 21862v2 Announce Type: replace-cross Abstract: Chunked vision-language-action (VLA) policies predict multi-step robot controls, conditioning each update on the current visual observation alone.
arXiv Machine Learning
Sep 16
TEMPO: Learning Temporal Context for Dynamic Robot Manipulation
arXiv:2609.16864v1 Announce Type: cross Abstract: Vision-language-action (VLA) models have achieved impressive performance in quasi-static manipulation, but struggle in dynamic manipulation tasks bec...
arXiv Machine Learning
Jun 11
Learning What to Say to Your VLA: Mostly Harmless Vision Language Action Model Steering
arXiv:2606. 12299v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models provide a natural language interface to robot control, but the mapping from language to behavior is often brittle and unintuitive: semantically similar instructions can induce drastically different behaviors, while some capabilities may not be elicitable through prompting alone.
arXiv AI
Aug 3
FBFM: A Training-Free Asynchronous Feedback Mechanism for Flow-Matching in World-Action Models Execution
arXiv:2607. 29235v1 Announce Type: cross Abstract: Although world-action models (WAMs) enhance long-horizon robot control by predicting visual evolution before acting, long-horizon reliability demands repeated re-grounding in real observations--not recursive rollout.
arXiv AI
Aug 18
Algorithm-Architecture Co-Design for Efficient VLA Inference via Speculative Inference and Verification
arXiv:2608. 15636v1 Announce Type: cross Abstract: Vision-Language-Action (VLA) models have demonstrated remarkable capabilities in the field of embodied AI, but their high computational cost and limited predicted action length hinder real-time deployment.
arXiv AI
Aug 19
LoopVLA: Learning Sufficiency in Recurrent Refinement for Vision-Language-Action Models
LoopVLA introduces a recurrent Vision‑Language‑Action architecture that learns to refine multimodal representations, predict actions, and estimate when further refinement is unnecessary. By iteratively applying a shared Transformer block and producing a sufficiency score at each step, it decouples refinement from fixed layer indices and aligns confidence scores with action quality through a self‑supervised objective. Experiments on LIBERO, LIBERO‑Plus, and VLA‑Arena demonstrate that LoopVLA reduces model parameters by 45% and boosts inference throughput up to 1.7× while matching or surpassing strong baselines in task success.
arXiv Computer Vision
Aug 27
StreamPI: Streaming Multimodal Temporal Modeling for Vision-Language-Action Models
StreamPI introduces a streaming multimodal temporal modeling framework that enhances Vision‑Language‑Action models by adding temporal reasoning without extra parameters. It anchors each visual observation and language instruction pair as a temporal unit, using bidirectional attention for cross‑modal fusion and causal attention for autoregressive streaming inference. The method employs random‑interval streaming training to improve robustness and leverages the LLM backbone’s length extrapolation to inherit pretrained weights, achieving superior performance over pi0.5 on real‑robot and simulation tasks.
arXiv Machine Learning
4d ago
Reactive Real-Time Flow Policies via Asynchronous Distribution Alignment
arXiv:2609.36540v1 Announce Type: cross Abstract: Generalist robot policies such as vision-language-action models (VLAs) have achieved remarkable generalization, but their inference delays can confli...
arXiv Machine Learning
Sep 14
Dynin-Robotics: Omnimodal Unified Diffusion Vision-Language-Action Model
Dynin‑Robotics introduces an omnimodal masked‑diffusion backbone, Dynin‑Omni, that jointly represents language, visual observations, goals, and actions as discrete tokens. By conditioning on different spans, the same model learns action prediction, next‑observation prediction, goal‑state prediction, and trajectory‑to‑instruction reconstruction, enabling test‑time scaling through goal prediction and action‑candidate evaluation. The system, pretrained on 1.33 million trajectories from 48 Open X‑Embodiment datasets, achieves competitive performance on LIBERO, zero‑shot LIBERO‑Plus, and a 78.4 % success rate on a Franka Research 3 robot, while a block‑parallel implementation speeds up action decoding by up to 29.2×.