Hang Yu

ML-Embed: Inclusive and Efficient Embeddings for a Multilingual World

Ziyin Zhang, Zihan Liao, Hang Yu, et al.

TingIS: Real-time Risk Event Discovery from Noisy Customer Incidents at Enterprise Scale

Jun Wang, Ziyin Zhang, Rui Wang, et al.

F2LLM-v2: Inclusive, Performant, and Efficient Embeddings for a Multilingual World

Ziyin Zhang, Zihan Liao, Hang Yu, et al.

F2LLM-v2 is a new family of multilingual AI embedding models (ranging from small to large sizes) that can understand and work with over 200 languages, including many underserved languages that other models ignore. The researchers made these models more efficient and faster than previous versions while maintaining high performance, and they're releasing everything publicly to help other researchers build better AI systems.