method
active
method:model-knowledge-editing-romeModel (knowledge) editing / ROME
Precise knowledge modification; ROME has flaws.
Neighborhood — ranked by edge-count
Papers (1)
paper
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.
- Technique for modifying model knowledge or behavior via targeted interventions, e.g., ROME by Meng et al.
- A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
- Ability to surgically alter model behavior through direct parameter changes rather than activation interventions.
- Probability of data under the model, penalizing complexity and rewarding accuracy.
- Application enabled by VPD: direct manipulation of weight matrices for interpretable model modification.
- Primary test domain for manifold steering, including reasoning and ICL tasks
- When a single conversation involves different models, e.g., routing between standard and reasoning models
- Primary substrate for manifold steering experiments; demonstrates method on reasoning and in-context tasks.