claim
The creators of the Vectara hallucination leaderboard prefer model-based evaluation over human evaluation because it provides a repeatable process that can be shared with others, whereas human annotation processes are difficult to replicate and share beyond the process description and labels.
Authors
Sources
- vectara/hallucination-leaderboard - GitHub github.com via serper
Referenced by nodes (2)
- Vectara LLM Hallucination Leaderboard concept
- human evolution concept