E1: Cross-Domain Anchoring Demonstrations

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

• Cross-domain anchoringconcept0.856
  Observation that anchoring effects appear across text and vision modalities.
• Cross-domain anchoring demonstrates that UCCT's principles apply beyond textclaim0.774
  Claim of modality generality
• semantic anchoringconcept0.705
  The central idea that external structure binds latent patterns to desired targets.
• anchorconcept0.699
  External structure (in-context examples, retrieval, tuning) that biases latent pattern activation.
• anchoring strength Sconcept0.689
  Composite score S = ρd − dr − log k predicting anchoring success.
• Language models would achieve some notion of grounding in the visual domain even in the absence of cross-modal training data, because they share a common modality-agnostic representationhypothesis0.689
  Implication of PRH for language model visual grounding
• Commonsense reasoning shows uniform but weaker anchoring (S ≈ −2.15)finding0.688
  Task-specific comparison.
• Semantic-anchoring assumptionconcept0.685
  Assumption that structured inputs act as anchors biasing activation toward target patterns PT.