CARMA: Enhanced Compositionality in LLMs via Advanced Regularisation and Mutual Information Alignment

Documenti analoghi

• Emergence and Localisation of Semantic Role Circuits in LLMs
  di: Aljaafari, Nura, et al.
  Pubblicazione: (2025)
• Interpreting token compositionality in LLMs: A robustness analysis
  di: Aljaafari, Nura, et al.
  Pubblicazione: (2024)
• TRACE: Training and Inference-Time Interpretability Analysis for Language Models
  di: Aljaafari, Nura, et al.
  Pubblicazione: (2025)
• TRACE for Tracking the Emergence of Semantic Representations in Transformers
  di: Aljaafari, Nura, et al.
  Pubblicazione: (2025)
• The Mechanics of Conceptual Interpretation in GPT Models: Interpretative Insights
  di: Aljaafari, Nura, et al.
  Pubblicazione: (2024)