quote
active
quote:we-need-methods-that-can-help-a-user-identify-what-a-model-will-learn-from-their-training-data-and-subsequently-shape-what-it-is-allowed-to-learnwe need methods that can help a user identify what a model will learn from their training data, and, subsequently, shape what it is allowed to learn
Core goal statement.
Source paper
extracted_from(2026) · Leon Bergen · Usha Bhalla · Sidharth Baskaran · Max Loeffler +13
Neighborhood — ranked by edge-count
Papers (1)
paper
Frameworks (1)
framework
- Pipeline from preference data to hypotheses to interventions.
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.
- Future direction.
- Opening question.
- Are there model structures that account for memorization and generalization across large swaths of training data?question0.812Aggregate analysis motivation.
- Author interpretation of selectivity results showing DAS advantage diminishes when controlling for expressivity
- Can we transform a model's features into an inexpensive, dense signal for learning open-ended tasks?question0.807Motivating question.
- Motivating question.
- Opening question.
- Authors argue features are model properties because logit effects and ablations are consistent with feature interpretations