Statistical Estimation of Adversarial Risk in Large Language Models under Best-of-N Sampling

Documenti analoghi

• ReEval: Automatic Hallucination Evaluation for Retrieval-Augmented Large Language Models via Transferable Adversarial Attacks
  di: Yu, Xiaodong, et al.
  Pubblicazione: (2023)
• Do More Details Always Introduce More Hallucinations in LVLM-based Image Captioning?
  di: Feng, Mingqian, et al.
  Pubblicazione: (2024)
• Discover and Mitigate Multiple Biased Subgroups in Image Classifiers
  di: Zhang, Zeliang, et al.
  Pubblicazione: (2024)
• Regularized Best-of-N Sampling with Minimum Bayes Risk Objective for Language Model Alignment
  di: Jinnai, Yuu, et al.
  Pubblicazione: (2024)
• Inference-Aware Fine-Tuning for Best-of-N Sampling in Large Language Models
  di: Chow, Yinlam, et al.
  Pubblicazione: (2024)