Critical Lesson
"Critical Lesson" research focuses on extracting valuable insights and best practices from diverse applications of artificial intelligence and machine learning. Current efforts concentrate on improving model performance and reliability across various domains, employing techniques like deep learning, retrieval-augmented generation, and mixture-of-experts models, while also addressing challenges in data quality, explainability, and ethical considerations. These lessons learned are crucial for advancing AI development, enhancing the trustworthiness of AI systems, and improving the design of human-centered AI applications across fields ranging from healthcare to robotics.
Papers
The Lessons of Developing Process Reward Models in Mathematical Reasoning
Zhenru Zhang, Chujie Zheng, Yangzhen Wu, Beichen Zhang, Runji Lin, Bowen Yu, Dayiheng Liu, Jingren Zhou, Junyang Lin
Lessons From Red Teaming 100 Generative AI Products
Blake Bullwinkel, Amanda Minnich, Shiven Chawla, Gary Lopez, Martin Pouliot, Whitney Maxwell, Joris de Gruyter, Katherine Pratt, Saphir Qi, Nina Chikanov, Roman Lutz, Raja Sekhar Rao Dheekonda, Bolor-Erdene Jagdagdorj, Eugenia Kim, Justin Song, Keegan Hines, Daniel Jones, Giorgio Severi, Richard Lundeen, Sam Vaughan, Victoria Westerhoff, Pete Bryan, Ram Shankar Siva Kumar, Yonatan Zunger, Chang Kawaguchi, Mark Russinovich
Lessons From an App Update at Replika AI: Identity Discontinuity in Human-AI Relationships
Julian De Freitas, Noah Castelo, Ahmet Uguralp, Zeliha Uguralp
ChocoLlama: Lessons Learned From Teaching Llamas Dutch
Matthieu Meeus, Anthony Rathé, François Remy, Pieter Delobelle, Jens-Joris Decorte, Thomas Demeester