Attention Head Entropy of LLMs Predicts Answer Correctness

Documenti analoghi

• LieRE: Lie Rotational Positional Encodings
  di: Ostmeier, Sophie, et al.
  Pubblicazione: (2024)
• Foundation Models in Radiology: What, How, When, Why and Why Not
  di: Paschali, Magdalini, et al.
  Pubblicazione: (2024)
• TRoVe: Discovering Error-Inducing Static Feature Biases in Temporal Vision-Language Models
  di: Varma, Maya, et al.
  Pubblicazione: (2025)
• From Detection to Mitigation: Addressing Bias in Deep Learning Models for Chest X-Ray Diagnosis
  di: Mottez, Clemence, et al.
  Pubblicazione: (2025)
• MedVAL: Toward Expert-Level Medical Text Validation with Language Models
  di: Aali, Asad, et al.
  Pubblicazione: (2025)