claim
active
claim:conceptual-belief-spaces-unify-geometric-theories-of-conceptual-representation-with-probabilistic-theories-of-learningConceptual belief spaces unify geometric theories of conceptual representation with probabilistic theories of learning
Theoretical aim.
Source paper
extracted_from(2026) · Eric Bigelow · Raphaël Sarfati · Daniel Wurgaft · Owen Lewis +4
Neighborhood — ranked by edge-count
Papers (1)
paper
Questions (1)
question
- Future work.
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.
- Core proposal: LLM beliefs are distributions over a low-dimensional conceptual space; ICL is a trajectory through it.
- How do the theory and model generalize to non-binary concept spaces with multiple belief directions?question0.829Future work.
- Speculative interpretation of breakdown.
- Interpretive assertion: the same geometric structure (e.g. circular for days) appears identically in both internal activations and output probabilities.
- Future work.
- Finding (3).
- Geometric theory of concepts as convex regions in metric spaces, extended by the paper.
- Abstract's closing thesis.