5arXiv cs.AI (Artificial Intelligence)·
LUNA: LBS-free neural model for 3D human avatar animation from 2D controls
LUNA is a new neural animation model that bypasses Linear Blend Skinning (LBS) and parametric body models to animate 3D human avatars directly from 2D inputs such as images, keypoints, and sketches. A transformer-based motion regressor disentangles global rigid motion from local non-rigid dynamics, while hybrid supervision enables training on both fitted data and unlabeled in-the-wild video. The model achieves competitive visual fidelity to LBS-based approaches and demonstrates zero-shot cross-identity generalization across diverse driving modalities. The authors claim LUNA is the first end-to-end 3D animatable model supporting implicit 2D driving.
Related guides (1)
Related events (8)
VLK: Synthetic vision-language-kinematics supervision enables humanoid loco-manipulation from egocentric observations
Researchers introduce a pipeline that generates 48,000 paired vision-language-kinematics trajectories synthetically using 3D Gaussian Splatting to reconstruct indoor scenes, bypassing the need for expensive human-annotated robot data. A VLK policy trained on this data predicts whole-body kinematic trajectories from egocentric images and language instructions, which a whole-body tracker converts to physical actions. The approach is validated on a Unitree G1 humanoid performing navigation and object transport, demonstrating viable sim-to-real transfer for perception-based loco-manipulation.
NVlabs/ProtoMotions: GPU-accelerated framework for physically simulated humanoid motion learning
NVIDIA Labs released ProtoMotions, an open-source Python framework for GPU-accelerated simulation and reinforcement learning of physically simulated digital humans and humanoid robots. The project is trending on GitHub with 1,916 stars and 63 new stars today. It sits at the intersection of physics simulation, embodied AI, and robotics training infrastructure.
ARDY: NVIDIA's streaming autoregressive diffusion framework for real-time controllable 3D human motion generation
NVIDIA researchers introduce ARDY, a streaming motion generation framework that combines autoregressive transformers with diffusion-based denoising to produce high-fidelity 3D human motions in real-time. The system uses a hybrid representation pairing explicit root features with latent body embeddings, enabling online text prompting and flexible long-horizon kinematic constraints simultaneously. ARDY is evaluated on HumanML3D and a large-scale proprietary dataset (Bones Rigplay), with an interactive demo showing dynamic text control, keyframe constraints, and locomotion. The work targets animation, simulation, and humanoid robotics applications where both controllability and inference speed are required.
Humanoid-GPT: GPT-style Transformer trained on 2B-frame motion corpus for zero-shot humanoid control
Researchers introduce Humanoid-GPT, a causal Transformer pre-trained on a 2-billion-frame retargeted motion corpus that unifies major mocap datasets with large-scale in-house recordings for whole-body humanoid control. The model achieves zero-shot generalization to unseen motions and control tasks, overcoming the agility-generalization trade-off seen in prior MLP-based trackers. Scaling analyses demonstrate a new performance frontier for dynamic motion tracking without task-specific fine-tuning.
LACUNA testbed introduces ground-truth parameter-level evaluation for LLM unlearning
Researchers introduce LACUNA, the first unlearning testbed with ground-truth parameter-level localization, designed to evaluate whether LLM unlearning methods truly erase knowledge from model weights or merely suppress it at the output level. The testbed injects PII of synthetic individuals into predefined parameters of 1B and 7B OLMo-based models via masked continual pretraining, enabling direct measurement of localization precision. Benchmarking current SOTA unlearning methods reveals they are highly imprecise and vulnerable to resurfacing attacks despite strong output-level performance, while successful localization enables even simple gradient-based methods to achieve robust erasure. The work addresses a critical gap in unlearning evaluation methodology relevant to privacy compliance and AI safety.
SmolVLA: Efficient Vision-Language-Action Model trained on Lerobot Community Data
Hugging Face introduces SmolVLA, a compact Vision-Language-Action model designed for robotics control, trained on community-contributed data from the LeRobot ecosystem. The model targets efficient deployment on resource-constrained hardware while maintaining competitive manipulation performance. This release represents a continuation of Hugging Face's strategy to democratize robotics AI through open community data pipelines.
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.
Welcome NVIDIA Cosmos 3: The First Open Omni-model for Physical AI Reasoning and Action
NVIDIA has released Cosmos 3, described as the first open omni-model targeting physical AI reasoning and action. The model is hosted and announced via Hugging Face, positioning it as an open-weights offering for robotics and embodied AI applications. The announcement highlights multimodal capabilities oriented toward physical world understanding and agent-level action.