Subjective Content Quality Assessment on GEO-bench
22.3Relevance ScoreGEO-SFE
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| GEO-SFEEvaluation Model=Gemini 2.5 Pro, Temperature=0.3, Normalization=Position-Adjusted Word Count distribution, Sample Size=2002026.03 | 22.3 | 23.1 | 23.6 | 22.8 | 23.4 | 22.6 | 23.7 | 23.1 | 18.5 | |
| BaselineEvaluation Model=Gemini 2.5 Pro, Temperature=0.3, Normalization=Position-Adjusted Word Count distribution, Sample Size=2002026.03 | 18.7 | 17.5 | 21.3 | 19.1 | 20.8 | 17.2 | 22.1 | 19.5 | — |