arXiv
InterEvolve: Test-Time Evolution of Reward Programs for Humanoid Loco-Manipulation
We study test-time evolution for humanoid loco-manipulation: solving tasks that a controller was nev...
HABERLER
Tüm Haberler
Yapay zekâ dünyasından son haberler, derinlikli analizler, teknik incelemeler.
114Yazı
1Yazar
5.1KOkuma · 30g
5Bu Hafta
arXiv
AssemblyWorld: Rethinking 3D Assembly with General-Purpose Agents
The task of 3D assembly requires translating an understanding of parts and their relationships into ...
arXiv
Ego4WAM: What Matters When Scaling Egocentric Human Data for Robot Learning?
Egocentric human data provides a scalable source of experience for robot learning, but varies substa...
arXiv
DynaHarness: A Dynamic Physical Harness for Self-Evolving Robot Agents
Pretrained robot policies provide useful action priors, but long-horizon manipulation still requires...
arXiv
Skill-Space Shooting for Autonomous Robot Policy Improvement
Robots deployed in the physical world must be able to improve beyond their initial training as they ...
arXiv
Counterfactual Video Generation Enables Scalable Humanoid Loco-Manipulation
Teaching humanoids loco-manipulation skills, such as carrying diverse objects, via visual imitation ...
arXiv
Rethinking Representations for World-Action Modeling
World-action models jointly learn robot policies and predict future observations, making the represe...
arXiv
Statistical Learning of Contractive Dynamical Representations for Composite Adaptive Control
We present a representation-learning framework for composite adaptive tracking control under dynamic...
arXiv
Systematic Multi-Agent Vision-and-Language Navigation: Formulation, Benchmark, and Method
Vision-and-Language Navigation (VLN) has largely focused on a single agent following a single instru...
arXiv
ENCP: Episode-Normalized Conformal Prediction for Vision-and-Language Navigation
Uncertainty estimation for Vision-Language-Navigation (VLN) models is a critical task since it can h...
arXiv
Çoklu Görev Manipülasyonu için Kendi Kendini Gösteren Üretken Kontrole Sahip VLA’ların İnce Ayarı
Pi0.5 gibi son teknoloji ürünü görüş-dil-eylem (VLA) modelleri, güçlü anlamsal anlayış, talimat taki...
arXiv
DriftWorld: Fast World Modeling through Drifting
Predictive world models enable robots to simulate the visual outcomes of their actions, but state-of...
arXiv
An offline approach to fNIRS-guided reinforcement learning for robot behavior
Human-in-the-loop Reinforcement Learning has become a popular approach for training, finetuning, and...
arXiv
GaP: A Graph-as-Policy Multi-Agent Self-Learning Harness For Variational Automation Tasks
For robots to work reliably in commercial and industrial applications, recent advances in agentic co...
arXiv
Provably Safe, Yet Scalable Reinforcement Learning
Safe reinforcement learning (RL) aims to learn policies that optimize rewards while satisfying const...