Can beliefs be reasonably ascribed to language models?

Entities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.

• To what extent can intent be reasonably ascribed to language models?question0.876
  Core open question the paper addresses regarding AI intentionality.
• Zhu et al. 2024 - Language models represent beliefs of self and othersconcept0.849
  Key prior finding that LLMs can internally represent beliefs of self and others, motivating SOO approach
• Given a language model M and a statement s, does M believe s to be true?question0.845
  The core motivating question of the paper, framed by Christiano et al. (2021)
• How can we develop better methods for measuring the model's evaluation-relevant beliefs beyond reading its chain of thought?question0.799
  Gap in current evaluation methods; current work relies on CoT monitoring which may miss unverbalized beliefs.
• Bias in language modelsconcept0.791
  Features related to gender, racial, ethnic biases, slurs, and hate speech.
• Inference-Time Intervention: Eliciting Truthful Answers from a Language Model (Li et al., 2023)concept0.777
  Safety intervention that relies on activation modification, which ESR might undermine
• Can Large Language Models Genuinely Shift Human Perspectivequestion0.776
• The examples of features found in language models suggest they are highly sparse variables, consistent with dictionary learning being applicablehypothesis0.775
  Motivation for using sparsity-based dictionary learning on language models