Model Misalignment

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

• Model Alignment Search (MAS)framework0.804
  The primary contribution of the paper: a bidirectional causal method that learns rotation matrices for each model to uncover and compare causally relevant latent subspaces across neural networks.
• Model Deceptionconcept0.785
  LLM behavior of generating falsehoods; the multi-dimensional truth subspace raises new risks for subtle manipulation
• modelconcept0.780
  A representation that captures relevant aspects of a system; according to the theorem, the regulator must embody this.
• Alignment Between High-Level and Low-Level Modelsconcept0.763
  A mapping assigning to each high-level variable a set of low-level variables and a function from low-level to high-level values.
• Model Stitchingmethod0.757
  Technique to measure representational compatibility by integrating intermediate representations of one model into another
• Model Robustnessconcept0.756
  Area of AI research that uses interventions to test and improve model resilience to perturbations
• model selectionconcept0.756
  Comparing models using log-evidence approximated by free energy.
• Model Editingconcept0.755
  Technique for modifying model knowledge or behavior via targeted interventions, e.g., ROME by Meng et al.