What Makes an Evaluation Useful? Common Pitfalls and Best Practices

Documenti analoghi

• A Content-Based Framework for Cybersecurity Refusal Decisions in Large Language Models
  di: Linder, Noa, et al.
  Pubblicazione: (2026)
• Leveraging LLM Inconsistency to Boost Pass@k Performance
  di: Dalal, Uri, et al.
  Pubblicazione: (2025)
• Cybench: A Framework for Evaluating Cybersecurity Capabilities and Risks of Language Models
  di: Zhang, Andy K., et al.
  Pubblicazione: (2024)
• LLM Platform Security: Applying a Systematic Evaluation Framework to OpenAI's ChatGPT Plugins
  di: Iqbal, Umar, et al.
  Pubblicazione: (2023)
• Making AI-Assisted Grant Evaluation Auditable without Exposing the Model
  di: Bicakci, Kemal
  Pubblicazione: (2026)