question
active
question:how-do-conceptual-belief-spaces-shape-uncertainty-in-long-form-text-generationHow do conceptual belief spaces shape uncertainty in long-form text generation?
Future work.
Source paper
extracted_from(2026) · Eric Bigelow · Raphaël Sarfati · Daniel Wurgaft · Owen Lewis +4
Neighborhood — ranked by edge-count
Papers (1)
paper
- Stories in Space: In-Context Learning Trajectories in Conceptual Belief Spaceassociated_withmentions
Claims (1)
claim
- Central thesis.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Theoretical aim.
- What is the right statistical model for explaining uncertainty dynamics in text generation?question0.763Central scientific question.
- Core proposal: LLM beliefs are distributions over a low-dimensional conceptual space; ICL is a trajectory through it.
- How do conceptual spaces relate to behaviors involving likelihood p(c | x) and predictive distributions p(y | c)?question0.744Future work.
- How do the theory and model generalize to non-binary concept spaces with multiple belief directions?question0.743Future work.
- Overall conclusion.
- Core open question the paper addresses through a behaviourist operationalisation.
- Speculative interpretation of breakdown.