A Head to Predict and a Head to Question: Pre-trained Uncertainty Quantification Heads for Hallucination Detection in LLM Outputs

Documenti analoghi

• Uncertainty-Aware Attention Heads: Efficient Unsupervised Uncertainty Quantification for LLMs
  di: Vazhentsev, Artem, et al.
  Pubblicazione: (2025)
• Faithfulness-Aware Uncertainty Quantification for Fact-Checking the Output of Retrieval Augmented Generation
  di: Fadeeva, Ekaterina, et al.
  Pubblicazione: (2025)
• Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph
  di: Vashurin, Roman, et al.
  Pubblicazione: (2024)
• Don't Throw Away Your Beams: Improving Consistency-based Uncertainties in LLMs via Beam Search
  di: Fadeeva, Ekaterina, et al.
  Pubblicazione: (2025)
• Unconditional Truthfulness: Learning Unconditional Uncertainty of Large Language Models
  di: Vazhentsev, Artem, et al.
  Pubblicazione: (2024)