Our Research Journey
We are actively pushing AI limits with a focus on safety. We invite you to explore the engines behind our solutions, and our contributions to the community.
Do-Not-Answer: A Dataset for Evaluating Safeguards in LLMs
Loki: An Open-source Tool for Fact Verification
M4: Multi-generator, Multi-domain, and Multi-lingual Black-Box Machine-Generated Text Detection
Balancing out Bias: Achieving Fairness Through Balanced Training
Against The Achilles' Heel: A Survey on Red Teaming for Generative Models
CMMLU: Measuring massive multitask language understanding in Chinese
FairLib: A unified framework for assessing and improving fairness
Keep up to date with LibrAI
Get the latest updates, research insights, and industry trends in AI safety and benchmarking delivered to your inbox. Stay informed with LibrAI.
