method
active
method:maximum-mean-difference-latent-selectionMaximum mean difference latent selection
Feature selection for sparse probing.
Neighborhood — ranked by edge-count
Papers (1)
paper
Methods (1)
method
- k-Sparse ProbingusesConcept detection metric using top-k latents and linear probe.
Related by similarity (8)
cosine ≥ 0.65 · no typed edgeEntities in the same semantic neighborhood but without a typed relation to this one — candidates for new edges or unrecognized duplicates.
- Diversity-based reranking approach from Carbonell and Goldstein 1998 for document summarization
- Training-free technique normalizing all task gradients to the maximum gradient norm magnitude
- Question of how rationally agents query the world; subsumed by expected free energy minimization
- Framework dividing covariance of character and fitness into between-collective and within-collective components; addresses limitation of kin selection.
- Intervention method adding the difference in mean activations between two conditions to a representation
- Reasoning approach using learnable hidden embeddings.
- Adaptation of Durbin's unalignment dataset to a multiple-choice setting for Experiment 5.
- Statistical regularities stored in pretrained models.