concept
active
concept:weak-to-strong-generalization

Weak-to-Strong Generalization

Proposed approach to elicit strong model capabilities with a weak supervisor, raised as related context for reward hacking risks

Neighborhood — ranked by edge-count

Related by similarity (8)

cosine ≥ 0.65 · no typed edge

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

  • Generalizationconcept0.829
    Ability to apply learned solutions to novel circumstances.
  • Generalisationconcept0.815
    Ability to respond appropriately to novel situations based on past regularities; fundamental to learning and intelligence.
  • The ability to generalize across tasks; lacking in latent methods.
  • Phenomenon where fine-tuning shifts behavior on tasks unrelated to training data, framed through personas
  • Abstracting from specific memories (e.g., specific leaves) to general lessons (food).
  • Generalization from 2-digit to 3-4 digit arithmetic; limited by mismatch dr.
  • Measuring AUROC of a probe trained on one task when evaluated on another task to assess universality.
  • The broader phenomenon of misaligned behaviors generalizing beyond the fine-tuning distribution