Theme of inquiry

When and why does chain-of-thought reasoning succeed or fail?

A question within its area, explored through 4 lines of inquiry below — each a family of specific questions the research asks.


What actually drives chain-of-thought reasoning improvements in language models?

43 specific questions

See all 43 questions in this line of inquiry
Do corrupted reasoning traces serve as effective supervision signals?

23 specific questions

See all 23 questions in this line of inquiry
Do reasoning traces faithfully represent or merely mimic actual model reasoning?

64 specific questions

See all 64 questions in this line of inquiry
Why do correct reasoning traces tend to be shorter than incorrect ones?

47 specific questions

See all 47 questions in this line of inquiry