Bigger models are more likely to converge to a shared representation than smaller models

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

• Different models cannot converge to the same representation if they have access to fundamentally different information; convergence is capped by mutual information between input signalsclaim0.853
  Key limitation of the PRH for non-bijective observations
• Scaling model size, as well as data and task diversity, drives representational convergence toward the platonic representationhypothesis0.834
  Core mechanism hypothesis connecting PRH to the empirical trend of scaling in AI
• There are fewer representations competent for N tasks than M<N tasks, so training more general models should yield fewer possible solutionshypothesis0.819
  Selective pressure toward convergence via task generality
• Larger models should amplify bias less than smaller models, with model biases more accurately reflecting data biases rather than exacerbating themclaim0.809
  Implication of PRH for AI fairness and bias
• How do representations differ or converge between architectures, tasks, and modalities?question0.808
  Broader research question MAS is positioned to address, citing multiple recent works.
• The model tends to reflect more when the question is difficult, and accuracy is generally lower for harder questionshypothesis0.797
  Hypothesis explaining negative correlation between reflection rate and accuracy without implying reflection is harmful
• Larger models can support higher-dimensional truth cones than smaller modelsclaim0.793
  Interpretation of ASR degradation patterns by model size across cone dimensions
• Interpretability features converge across different model architectures, revealing structural similarities.claim0.788