#multi-scale-learning News & Analysis

2 articles tagged with #multi-scale-learning. AI-curated summaries with sentiment analysis and key takeaways from 50+ sources.

2 articles

AINeutralarXiv – CS AI · Jun 255/10

🧠

HiT-JEPA: A Hierarchical Self-supervised Trajectory Embedding Framework for Similarity Computation

Researchers introduce HiT-JEPA, a hierarchical self-supervised learning framework that represents urban trajectory data across multiple semantic levels to improve similarity computation. The model captures fine-grained movement details, intermediate patterns, and high-level abstractions simultaneously, addressing limitations in existing approaches that struggle to balance local nuances with global dependencies.

AIBullisharXiv – CS AI · Jun 16/10

🧠

Mixture of Horizons in Action Chunking

Researchers propose Mixture of Horizons (MoH), a novel technique for vision-language-action models in robotics that processes action sequences at multiple time scales simultaneously to balance long-term planning with short-term precision. The method achieves state-of-the-art performance on robotic manipulation tasks, reaching 99% success rate on LIBERO benchmarks while enabling 2.5x faster inference through adaptive horizon selection.