[KongchangAI]
ConceptSelf-Preference Bias

Self-Preference

A systematic bias in which an LLM evaluator can recognize its own generated content and gives it preferential treatment, resulting in inflated self-evaluation scores.

Timeline (last 90 days)

Oct 19

Models tend to give high scores to their own outputs (or outputs of similar style); Panickssery et al. found that LLM evaluators can recognize their own generated content and favor it.

Unverified50%

All Facts (1)

Source Articles