← Все кластеры
Weak-to-strong generalization
closed
Тип событияfunding
Темаai safety
Организация
Страна
Статей2
Уник. источников1
Важность / Момент0.94 / 0
Период14.12.2023 00:00 — 14.12.2023 08:00
Создан06.04.2026 05:59:46
Статьи в кластере 2
Заголовок Источник Дата публикации Score
S Weak-to-strong generalization openai 14.12.2023 00:00 1
Embedding sim.1
Entity overlap1
Title sim.1
Time proximity1
NLP типother
NLP организация
NLP темаai safety
NLP страна

Открыть оригинал

We present a new research direction for superalignment, together with promising initial results: can we leverage the generalization properties of deep learning to control strong models with weak supervisors?
Superalignment Fast Grants openai 14.12.2023 08:00 0.71
Embedding sim.0.8109
Entity overlap0.2857
Title sim.0.0962
Time proximity0.9524
NLP типfunding
NLP организация
NLP темаai safety
NLP страна

Открыть оригинал

We’re launching $10M in grants to support technical research towards the alignment and safety of superhuman AI systems, including weak-to-strong generalization, interpretability, scalable oversight, and more.