Self-Improvement
التحسين الذاتي
نموذج تدريبي يُحسِّن فيه الذكاء الاصطناعي مخرجاته بنقدها وتنقيحها بنفسه وفقاً لمبادئ محددة، بدلاً من الاعتماد كلياً على تغذية راجعة خارجية من مُقيِّمين بشريين.
A training paradigm where the AI improves its own outputs by critiquing and revising them according to specified principles, rather than relying entirely on external feedback from human evaluators.
Also translated asالتحسُّن الذاتي، التعلُّم الذاتي التصحيحي
First appears in this corpus in: A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence (1956)
Appears in these papers
- Constitutional AI: Harmlessness from AI Feedback2022in the sky ✦
- Constitutional AI: Harmlessness from AI Feedback2022in the sky ✦
- A Proposal for the Dartmouth Summer Research Project on Artificial Intelligence1956in the sky ✦
- RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback2023in the sky ✦
- RLAIF: Scaling Reinforcement Learning from Human Feedback with AI Feedback2023in the sky ✦
- STaR: Bootstrapping Reasoning With Reasoning2022in the sky ✦