Do different prompting methods yield a common task representation in language models?

Documenti analoghi

• Compositional learning of functions in humans and machines
  di: Zhou, Yanli, et al.
  Pubblicazione: (2024)
• A systematic investigation of learnability from single child linguistic input
  di: Qin, Yulu, et al.
  Pubblicazione: (2024)
• An explainable transformer circuit for compositional generalization
  di: Tang, Cheng, et al.
  Pubblicazione: (2025)
• Goals as Reward-Producing Programs
  di: Davidson, Guy, et al.
  Pubblicazione: (2024)
• Are they human? Detecting large language models by probing human memory constraints
  di: Schug, Simon, et al.
  Pubblicazione: (2026)