Projection Difference

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

• Sampling-Based Approximation of Projection Differencemethod0.835
  Efficient estimation strategy for projection difference using a random subset of training data to reduce computational cost
• Prompt Token Approximation of Projection Differencemethod0.787
  Uses last prompt token projection to approximate base generation projection, avoiding expensive model rollouts
• Projection difference is more predictive of post-finetuning trait behavior than raw projection of training datafinding0.769
  Justifies the use of projection difference metric rather than simpler raw projection for data screening
• Projection difference better captures training data influence on persona than raw projection because it accounts for the base model's natural response tendenciesclaim0.739
  Author's interpretive explanation for why projection difference outperforms raw projection in data screening
• concept representationconcept0.727
  How a neural network encodes a semantic concept internally, argued to be better captured by manifolds than by atomic features.
• Differentiating Space Proceduremethod0.720
  A layout method where objects are shaped by subdividing the space to fit, rather than arranging fixed modules.
• Perceptionconcept0.718
  Equated with inference of past, present and future hidden states via minimization of variational free energy.
• Position Encodingsconcept0.716
  Mechanism for encoding sequence order in transformers; paper argues these should reflect learned structural representations rather than fixed sines/cosines.