RESEARCH · CAT. R1
Research
Pushing the boundaries of artificial intelligence through rigorous research and open collaboration.
FOCUS AREAS · 001
PL. I
Large Language Models
Advancing language models through novel architectures, training techniques, and alignment methods.
PL. II
Multimodal Learning
Systems that understand and generate across text, images, and audio.
PL. III
AI Safety & Alignment
Ensuring systems remain aligned with human values as they grow more capable.
PL. IV
Reinforcement Learning
Agents that learn complex behaviours through interaction with their environment.
PUBLICATIONS · 002
Scaling Laws for Neural Language Models
Constitutional AI: Harmlessness from AI Feedback
Efficient Transformer Architectures for Long-Context Tasks
Multimodal Reasoning in Large Models
OPEN RESEARCH · 003
Open
Research
We believe the biggest breakthroughs come from collaboration. We actively publish our research, share our findings, and work with the global AI community.
Collaborate With Us50+
100+
25+
10+