The Mechanism of Weak-to-Strong Generalization: Feature Elicitation from Latent Knowledge

Documenti analoghi

• Transformers Learn Nonlinear Features In Context: Nonconvex Mean-field Dynamics on the Attention Landscape
  di: Kim, Juno, et al.
  Pubblicazione: (2024)
• In-Context Learning Is Provably Bayesian Inference: A Generalization Theory for Meta-Learning
  di: Wakayama, Tomoya, et al.
  Pubblicazione: (2025)
• Degrees of Freedom for Linear Attention: Distilling Softmax Attention with Optimal Feature Efficiency
  di: Nishikawa, Naoki, et al.
  Pubblicazione: (2025)
• Mamba Can Learn Low-Dimensional Targets In-Context via Test-Time Feature Learning
  di: Oh, Junsoo, et al.
  Pubblicazione: (2025)
• Weak-to-Strong Generalization with Failure Trajectories: A Tree-based Approach to Elicit Optimal Policy in Strong Models
  di: Ye, Ruimeng, et al.
  Pubblicazione: (2025)