Copertina

KERV: Kinematic-Rectified Speculative Decoding for Embodied VLA Models

Fuente: arXiv

Salvato in:

Dettagli Bibliografici
Autori principali:	Zheng, Zihao, Mao, Zhihao, Li, Maoliang, Chen, Jiayu, Sun, Xinhao, Zhang, Zhaobo, Cao, Donggang, Mei, Hong, Chen, Xiang
Natura:	Preprint
Pubblicazione:	2026
Soggetti:	Robotics Machine Learning
Accesso online:	Acceder al recurso
Tags:	Aggiungi Tag Nessun Tag, puoi essere il primo ad aggiungerne!!

Documenti analoghi

HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

DyQ-VLA: Temporal-Dynamic-Aware Quantization for Embodied Vision-Language-Action Models
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

VLN-Cache: Enabling Token Caching for VLN Models with Visual/Semantic Dynamics Awareness
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

RoboECC: Multi-Factor-Aware Edge-Cloud Collaborative Deployment for VLA Models
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

RAPID: Redundancy-Aware and Compatibility-Optimal Edge-Cloud Partitioned Inference for Diverse VLA Models
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

FreqCache: Accelerating Embodied VLN Models with Adaptive Frequency-Guided Token Caching
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

EaqVLA: Encoding-aligned Quantization for Vision-Language-Action Models
di: Jiang, Feng, et al.
Pubblicazione: (2025)

Anticipation-VLA: Solving Long-Horizon Embodied Tasks via Anticipation-based Subgoal Generation
di: Zhang, Zhilong, et al.
Pubblicazione: (2026)

Vidarc: Embodied Video Diffusion Model for Closed-loop Control
di: Feng, Yao, et al.
Pubblicazione: (2025)

SimVLA: A Simple VLA Baseline for Robotic Manipulation
di: Luo, Yuankai, et al.
Pubblicazione: (2026)

2D or 3D: Who Governs Salience in VLA Models? -- Tri-Stage Token Pruning Framework with Modality Salience Awareness
di: Zheng, Zihao, et al.
Pubblicazione: (2026)

AIM: Intent-Aware Unified world action Modeling with Spatial Value Maps
di: Fan, Liaoyuan, et al.
Pubblicazione: (2026)

STaRR: Spatial-Temporal Token-Dynamics-Aware Responsive Remasking for Diffusion Language Models
di: Sun, Xinhao, et al.
Pubblicazione: (2025)

Can VLA Models Learn from Real-World Data Continually without Forgetting?
di: Zhu, Jiarun, et al.
Pubblicazione: (2026)

DynaMo: Runtime Switchable Quantization for MoE with Cross-Dataset Adaptation
di: Zheng, Zihao, et al.
Pubblicazione: (2025)

TrackVLA++: Unleashing Reasoning and Memory Capabilities in VLA Models for Embodied Visual Tracking
di: Liu, Jiahang, et al.
Pubblicazione: (2025)

AtomVLA: Scalable Post-Training for Robotic Manipulation via Predictive Latent World Models
di: Sun, Xiaoquan, et al.
Pubblicazione: (2026)

Embodied Red Teaming for Auditing Robotic Foundation Models
di: Karnik, Sathwik, et al.
Pubblicazione: (2024)

AsyncVLA: An Asynchronous VLA for Fast and Robust Navigation on the Edge
di: Hirose, Noriaki, et al.
Pubblicazione: (2026)

Discrete Diffusion VLA: Bringing Discrete Diffusion to Action Decoding in Vision-Language-Action Policies
di: Liang, Zhixuan, et al.
Pubblicazione: (2025)

VLA-Touch: Enhancing Vision-Language-Action Models with Dual-Level Tactile Feedback
di: Bi, Jianxin, et al.
Pubblicazione: (2025)

From Inference Efficiency to Embodied Efficiency: Revisiting Efficiency Metrics for Vision-Language-Action Models
di: Li, Zhuofan, et al.
Pubblicazione: (2026)

Embodied-R1: Reinforced Embodied Reasoning for General Robotic Manipulation
di: Yuan, Yifu, et al.
Pubblicazione: (2025)

NaviMaster: Learning a Unified Policy for GUI and Embodied Navigation Tasks
di: Luo, Zhihao, et al.
Pubblicazione: (2025)

Spatial-Temporal Graph Diffusion Policy with Kinematic Modeling for Bimanual Robotic Manipulation
di: Lv, Qi, et al.
Pubblicazione: (2025)

MetaVLA: Unified Meta Co-training For Efficient Embodied Adaption
di: Li, Chen, et al.
Pubblicazione: (2025)

D-CAT: Decoupled Cross-Attention Transfer between Sensor Modalities for Unimodal Inference
di: Daher, Leen, et al.
Pubblicazione: (2025)

Self-Improving Embodied Foundation Models
di: Ghasemipour, Seyed Kamyar Seyed, et al.
Pubblicazione: (2025)

OpenVLA: An Open-Source Vision-Language-Action Model
di: Kim, Moo Jin, et al.
Pubblicazione: (2024)

CEED-VLA: Consistency Vision-Language-Action Model with Early-Exit Decoding
di: Song, Wenxuan, et al.
Pubblicazione: (2025)

Shifting Uncertainty to Critical Moments: Towards Reliable Uncertainty Quantification for VLA Model
di: Tang, Yanchuan, et al.
Pubblicazione: (2026)

Vidar: Embodied Video Diffusion Model for Generalist Manipulation
di: Feng, Yao, et al.
Pubblicazione: (2025)

DropVLA: An Action-Level Backdoor Attack on Vision-Language-Action Models
di: Xu, Zonghuan, et al.
Pubblicazione: (2025)

HyperVLA: Efficient Inference in Vision-Language-Action Models via Hypernetworks
di: Xiong, Zheng, et al.
Pubblicazione: (2025)

UAV-Track VLA: Embodied Aerial Tracking via Vision-Language-Action Models
di: Zhang, Qiyao, et al.
Pubblicazione: (2026)

Don't Run with Scissors: Pruning Breaks VLA Models but They Can Be Recovered
di: Jabbour, Jason, et al.
Pubblicazione: (2025)

Robotic Control via Embodied Chain-of-Thought Reasoning
di: Zawalski, Michał, et al.
Pubblicazione: (2024)

Model Adaptation for Time Constrained Embodied Control
di: Song, Jaehyun, et al.
Pubblicazione: (2024)

VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model
di: Li, Wenhao, et al.
Pubblicazione: (2026)

SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning
di: Li, Haozhan, et al.
Pubblicazione: (2025)