Lingchu AI demonstrates that scaling human‑operation data to nearly 100,000 hours, combined with a two‑model system and reinforcement learning, can replace costly robot‑teleoperation data and achieve top performance on the MolmoSpaces benchmark.
HumanRig introduces a large‑scale dataset of 11,434 AI‑generated T‑pose humanoid meshes with unified skeletons, skinning weights, joint data and images, and leverages it in a novel automatic rigging pipeline—featuring a prior‑guided skeleton estimator, a U‑shaped point transformer, and a mesh‑skeleton mutual attention network—that significantly outperforms previous methods in skeleton accuracy and skinning quality.
This article presents Ant Group's multimodal research on video retrieval, detailing a large Chinese video‑text pre‑training dataset, three techniques that raise video‑text semantic search performance by up to 24.5%, and an end‑to‑end video‑video copyright detection system that reduces storage by 85% and speeds up inference 18‑fold.
