Aligned-MTL

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

• Alignmentconcept0.771
  The goal of making model behavior match human values and intentions, often addressed during post-training.
• Aligned by Designconcept0.765
  Paper's proposed strategy of instilling intrinsic moral cognition so AI remains aligned even as capabilities expand
• Alignment Map (ϕ)concept0.751
  The bijective function mapping DNN inner neurons to latent variables in causal abstraction; its complexity is the central variable studied
• Nash-MTLmethod0.738
  Gradient aggregation via Nash bargaining game.
• Model Alignment Search (MAS)framework0.737
  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.
• Cross-Modal Alignmentconcept0.734
  The alignment between representations learned from different data modalities such as vision and language
• DB-MTL (Dual-Balancing Multi-Task Learning)method0.732
  The proposed method combining loss-scale balancing via logarithm transformation and gradient-magnitude balancing via maximum-norm normalization.
• Representational Alignmentconcept0.732
  Measure of similarity between the similarity structures (kernels) induced by two different representations