Jinyi Bai
Titles as the sources gave them. In 1 entry, with 0 checked quotations.
In the Base
Every entry they appear in.
Trustworthy Recursive Self-Improvement in the Era of Large Models: Foundation, Risks, Evaluation and Open Challenges
According to the abstract, Pu, Gao, Bai and colleagues survey recursive self-improvement in large-model systems, giving an operational definition and process framework. They organize requirements for trust into four themes, including constraint preservation, security and human oversight.