Humans prefer representative distributions over non-representative ones when evaluating outputs at the distribution level, providing a reward gap that VS exploits

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

• In active inference, reward can simply be treated as another observation we have a preference over, rather than a special signal.claim0.787
  Abstract; central distinction.
• The results generalise readily to non-equilibrium systems where scaling relationships remain similar (e.g., dynamic or localised scaling).claim0.787
  Claim about broader applicability of the scaling argument
• Different prompt types collapse to different modes: instance prompts collapse to a single prototypical response, while distribution prompts collapse to a representative high-entropy sampleclaim0.776
  The theoretical mechanism explaining why VS works despite mode collapse remaining operative
• Base models assign higher likelihood to typical-set (representative) sequences than to degenerate sequences under VS promptshypothesis0.768
  Assumption D.6 formalized in the theoretical framework; empirically validated with coin-flip typicality rating experiments
• Human preference voting measures popularity, not phenomenology or what mode a model was in when producing output.claim0.768
• A linear preference vector encoding how much a model likes a given task is persona-relative: it activates strongly for creative writing in the assistant but for phishing in the evil personafinding0.765
  Evidence that core representations like preferences are persona-relative, supporting claim that personas gate content of representations
• Evolution often produces general-purpose problem-solving machines whose capacities cannot be inferred from the default invariant course of development.claim0.764
  A claim about the outcome of the MCA-enhanced process.
• The claim that human-sized minds are optimal for generating welfare per resource unit may seem a little suspicious absent further justification, since the scaling relationship is unlikely to have a peak at human mind sizeclaim0.764
  Supports the mind-scale dimension of super-beneficiary status