News

INFLY TECH Team Solves the Problem of Diversity Collapse in Large Model Training

This groundbreaking research, jointly completed by INFLY TECH, Fudan University, and Griffith University, was published in ...

INFLY TECH DPH-RL Framework: A New Breakthrough in AI Reinforcement Learning, Saying Goodbye to the Dilemma of 'Specialization' in Large Model Training

The DPH-RL framework recently released by the INFLY TECH team brings a new breakthrough to the field of artificial intelligence. This framework aims to address the paradox of 'training more leads to ...