Theme of inquiry

How do contextual factors, biases, and expectations shape LLM outputs?

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


Do language models reason like humans or mimic surface patterns?

96 specific questions

See all 96 questions in this line of inquiry
How do LLM judges' systematic biases affect alignment and evaluation outcomes?

44 specific questions

See all 44 questions in this line of inquiry
Is language model reasoning authentic and what causes models to reason?

76 specific questions

See all 76 questions in this line of inquiry
Do language models possess genuine introspective self-awareness or only behavioral mimicry?

23 specific questions

See all 23 questions in this line of inquiry