concept
active
concept:cross-architecture-generalizationCross-Architecture Generalization
Whether learned cones transfer effectively across model families (Qwen vs Gemma) and sizes
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.
- Measuring AUROC of a probe trained on one task when evaluated on another task to assess universality.
- Ability to apply learned solutions to novel circumstances.
- Ability to respond appropriately to novel situations based on past regularities; fundamental to learning and intelligence.
- Generalization from 2-digit to 3-4 digit arithmetic; limited by mismatch dr.
- Explicit textual or graphical links between parts of a work, dynamic and virtual.
- Abstracting from specific memories (e.g., specific leaves) to general lessons (food).
- Validation of judge model robustness by regrading 1000 responses with 4 additional judge models
- The ability to generalize across tasks; lacking in latent methods.