However, most alignment research is not very crisp and requires research taste when evaluating. Thi...
- 对齐研究通常缺乏明确性
- 可扩展监督是关键挑战
- 人类只能提供弱监督
This is why we chose to point the AAR at this scalable oversight problem! Progress would let AARs work on fuzzier alignment problems, where humans can only provide weak supervision." / X
Jan Leike on X: "However, most alignment research is not very crisp and requires research taste when evaluating. This is why we chose to point the AAR at this scalable oversight problem! Progress would let AARs work on fuzzier alignment problems, where humans can only provide weak supervision." / X
Don’t miss what’s happening
People on X are the first to know.
Post
See new posts
Conversation
However, most alignment research is not very crisp and requires research taste when evaluating. This is why we chose to point the AAR at this scalable oversight problem! Progress would let AARs work on fuzzier alignment problems, where humans can only provide weak supervision.
·
2
3
57
3
New to X?
Sign up now to get your own personalized timeline!
Sign up with Apple
By signing up, you agree to the Terms of Service and Privacy Policy, including Cookie Use.
Relevant people
-  Jan Leike @janleike Follow Click to Follow janleike Alignment team lead @AnthropicAI . Previously OpenAI & DeepMind. Optimizing for a post-AGI future where humanity flourishes. Opinions aren't my employer's.
Trending now
What’s happening
Trending worldwide
23 Nisan Ulusal Egemenlik
Trending worldwide
Ateneo
Trending worldwide
Pay希望
Music · Trending
#TXT29thWin
|
|
|
|
|
More
© 2026 X Corp.