1

Abstract Acquiring physically plausible motor skills across diverse and unconventional morphologies—from humanoids to ants—is crucial for robotics and simulation. We introduce No-data Imitation Learning (NIL), which: Generates a reference video with a pretrained video diffusion model from a single simulation frame + text prompt.

Abstract We present HaPTIC, an approach that infers coherent 4D hand trajectories from monocular videos. Current video-based hand pose reconstruction methods primarily focus on improving frame-wise 3D pose using adjacent frames rather than studying consistent 4D hand trajectories in space.

Abstract Estimating sewing patterns from images is a practical approach for creating high-quality 3D garments. Due to the lack of real-world pattern-image paired data, prior approaches fine-tune large vision-language models (VLMs) on synthetic garment datasets, limiting their generalization to real-world images.

Abstract Hands play a central role in daily life, yet modeling natural hand motions remains underexplored. Existing methods that tackle text-to-hand-motion generation or hand animation captioning rely on studio-captured datasets with limited actions and contexts, making them costly to scale to in-the-wild settings.

Abstract Hands are central to interacting with our surroundings and conveying gestures, making their inclusion essential for full-body motion synthesis. Yet existing methods either ignore hands entirely or operate under highly constrained conditions.

Abstract Human-like motion requires human-like perception. We introduce CLOPS, the first human avatar that solely uses egocentric vision to perceive its surroundings and navigate 3D scenes. Rather than relying on task-specific perception methods or privileged state information, CLOPS argues that human-like avatar behavior requires human-like perception.

Abstract While current general-purpose 3D human models (e.g., SMPL-X) efficiently represent accurate human shape and pose, they lack the ability to physically interact with the environment due to their kinematic nature.

Abstract We introduce InteractVLM, a novel method to estimate 3D contact points on human bodies and objects from single in-the-wild images, enabling accurate human-object joint reconstruction in 3D. This is challenging due to occlusions, depth ambiguities, and widely varying object shapes.

Abstract Benchmarks are essential for tracking rapid LLM progress—but today’s models exceed 90% on tasks like MMLU, saturating existing exams. We introduce Humanity’s Last Exam (HLE), a multi-modal, closed-ended benchmark spanning 2,500 questions across 100+ subjects at the frontier of human knowledge.

Abstract Synthesizing accurate hand–object interactions (HOI) is critical for AR/VR and vision tasks. Existing dense–correspondence methods improve contact fidelity but lack full differentiability or generality. We propose CHOIR, a versatile, fully differentiable interaction field: