Theme of inquiry

What enables reasoning capability in transformer language models?

A question within its area, explored through 6 lines of inquiry below — each a family of specific questions the research asks.


Can models improve accuracy without degrading reasoning quality?

69 specific questions

See all 69 questions in this line of inquiry
What reasoning architectures enable models to solve complex problems efficiently?

51 specific questions

See all 51 questions in this line of inquiry
Is reasoning capability latent in base models or created by post-training?

74 specific questions

See all 74 questions in this line of inquiry
What types of diversity prevent reasoning systems from collapsing?

39 specific questions

See all 39 questions in this line of inquiry
How much does training format versus domain influence reasoning?

19 specific questions

See all 19 questions in this line of inquiry
Can reasoning scale in latent space without tokens?

40 specific questions

See all 40 questions in this line of inquiry