hypothesis
active
hypothesis:asymmetry-between-positive-and-negative-interventions-is-explained-by-self-repair-restoring-ablated-featuresAsymmetry between positive and negative interventions is explained by self-repair restoring ablated features
Possible explanation.
Source paper
extracted_fromNeighborhood — ranked by edge-count
Papers (1)
paper
Findings (1)
finding
- Asymmetry in steering on echo features.
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.
- Gradient-based attribution approximates ablation impact, enabling fast search for causally important features.
- The core testable hypothesis driving the experimental design
- DiffLogic CA circuits exhibited emergent fault tolerance without explicit design for it
- Framework by Lee et al. explaining self-correction via linear latent concept directions, closely related prior work.
- Demonstrates the self-prior captures visual-proprioceptive associations, functioning as a probabilistic body schema
- Key interpretive conclusion from the dissociation between attempt rate and improvement rate in fine-tuning experiments
- Theoretical interpretation linking the self-prior to the established body schema concept