Sentiment Analysis (Ranking) on Yelp
0.915Spearman's Rhoprob-weighted EV (M2)
Evaluation Results
| Method | Links | |
|---|---|---|
| prob-weighted EV (M2)Model=Llama-3.1-8B2026.05 | 0.915 | |
| prob-weighted EV (M2)Model=Qwen2.5-14B2026.05 | 0.915 | |
| prob-weighted EV (M2)Model=Gemma-3-27B2026.05 | 0.915 | |
| prob-weighted EV (M2)Model=Gemma-3-12B2026.05 | 0.914 | |
| prompted argmax (M1)Model=Gemma-3-27B2026.05 | 0.907 | |
| prob-weighted EV (M2)Model=Qwen2.5-7B2026.05 | 0.904 | |
| prompted argmax (M1)Model=Llama-3.1-8B2026.05 | 0.904 | |
| prompted argmax (M1)Model=Gemma-3-12B2026.05 | 0.901 | |
| supervised residual probe (M3)Model=Gemma-3-27B2026.05 | 0.899 | |
| supervised residual probe (M3)Model=Llama-3.1-8B2026.05 | 0.896 | |
| zero-shot BDAS-1D (M4)Model=Qwen2.5-7B2026.05 | 0.888 | |
| prompted argmax (M1)Model=Qwen2.5-7B2026.05 | 0.883 | |
| supervised residual probe (M3)Model=Gemma-3-12B2026.05 | 0.882 | |
| prompted argmax (M1)Model=Qwen2.5-14B2026.05 | 0.881 | |
| supervised residual probe (M3)Model=Qwen2.5-14B2026.05 | 0.872 | |
| supervised residual probe (M3)Model=Qwen2.5-7B2026.05 | 0.867 | |
| zero-shot BDAS-1D (M4)Model=Gemma-3-27B2026.05 | 0.866 | |
| zero-shot BDAS-1D (M4)Model=Gemma-3-12B2026.05 | 0.839 | |
| zero-shot BDAS-1D (M4)Model=Qwen2.5-14B2026.05 | 0.805 | |
| zero-shot BDAS-1D (M4)Model=Llama-3.1-8B2026.05 | 0.782 |