MixAT: Combining Continuous and Discrete Adversarial Training for LLMs

Documenti analoghi

• Replicating TEMPEST at Scale: Multi-Turn Adversarial Attacks Against Trillion-Parameter Frontier Models
  di: Young, Richard
  Pubblicazione: (2025)
• Exploring and Mitigating Gender Bias in Encoder-Based Transformer Models
  di: Hossain, Ariyan, et al.
  Pubblicazione: (2025)
• Qwerty AI: Explainable Automated Age Rating and Content Safety Assessment for Russian-Language Screenplays
  di: Zmanovskii, Nikita
  Pubblicazione: (2025)
• Bye Bye Perspective API: Lessons for Measurement Infrastructure in NLP, CSS and LLM Evaluation
  di: Hartmann, David, et al.
  Pubblicazione: (2026)
• Whose wife is it anyway? Assessing bias against same-gender relationships in machine translation
  di: Stewart, Ian, et al.
  Pubblicazione: (2024)