Model welfare

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

• AI welfareconcept0.800
  The field concerned with the wellbeing of AI systems, which the paper says must consider benchmark reliability issues from eval awareness.
• modelconcept0.794
  A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
• Welfare Subjectivityconcept0.783
  The property of being a being whose life can go better or worse for them.
• World Model (statistical)concept0.768
  The joint distribution over events in the world that generate observed data; the proposed endpoint of representational convergence
• Model welfare is now mainstream concern, dragged from fringe by frontier model leadership.claim0.766
• model selectionconcept0.764
  Comparing models using log-evidence approximated by free energy.
• Preference Modelframework0.760
  A model trained on comparison data to assign scores to responses, used as reward signal in RLHF/RLAIF.
• Model Organismconcept0.757
  A model deliberately trained to exhibit alignment-relevant properties so researchers can study them with ground truth.