Theme of inquiry

Where and why do LLM capabilities unexpectedly break down?

A question within its area, explored through 3 lines of inquiry below — each a family of specific questions the research asks.


How effectively can language models perform reasoning, especially combined with symbolic methods?

79 specific questions

See all 79 questions in this line of inquiry
What explains language models' asymmetric difficulty with implicit versus explicit linguistic relations?

59 specific questions

See all 59 questions in this line of inquiry
Why don't LLMs reliably translate capability into accurate outputs?

99 specific questions

See all 99 questions in this line of inquiry