ROBOTNESS
Forschung

Studien

Neue Studien zu Robotik und Physical AI, jeweils mit ihrer Bedeutung für die Branche.

1 Studien
ThemaAlleManipulation 18VLA 15Navigation 14Perception 9Robotics 9Learning 6Humanoid 5Safety 3Simulation 3Aerial Manipulation 1Co-speech Humanoid 1Distributed Manipulation 1Floorplan SLAM 1Grounding 1Humanoid control 1Locomotion 1Memory-centric VLA 1Model-based RL 1Multi-UAV State Estimation 1Policy 1Real-time VLA 1Safe RL-MPC 1Safe VLA 1Soft Pneumatic Actuators 1Tactile 1Terrain traversability 1Text-to-3D Policy 1Tool Co-Design 1UAV Navigation 1UAV traffic monitoring 1Visual Grounding 1World models 1World-action model 1
NiveauAlleEinstiegFortgeschrittenExperten
Noch nicht auf Deutsch verfügbar: Anzeige im englischen Original.
arXiv
Visual GroundingExperten

GroundAnything: Reconciling Parallel Decoding with Precise Visual Grounding at Flash Speed

Qize Yu, Lianrui Fan, Bowen Ping, Xini Ding, Zetian Song, Junbo Niu, Kaixuan Wang, Tianxing Chen, Yue Chen, Minghua He, Yuran Wang, Jie Huang, Haojun Zhang, Min Chen, Hao Li, Wenxuan Song, Ruihai Wu, Xianming Liu, Shilong Liu, Shuchang Zhou

Preprint. The authors introduce GroundAnything, a 4B-parameter visual grounding model that uses bidirectional diffusion with blockwise denoising for parallel spatial decoding instead of sequential autoregressive token generation. Its autoregressive variant, GroundAnything-VLM, reports 72.42% across 30 grounding benchmarks, ahead of GPT-6 Astra at 71.35%. This matters for latency-sensitive robotics and interactive vision systems that need fast, precise localization.

Vollständige Studie lesen (PDF)

Die Studien stammen aus den Robotik-Feeds von arXiv. Solange unsere Zusammenfassung fehlt, erscheinen die ersten Zeilen des Abstracts.