Community Notes Generation on POLBENCH (test)
22.117Num URLs UsedGrok-4
Evaluation Results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Grok-4Type=Web Agent, Timestamp=T12026.02 | 22.117 | 0.167 | 0.857 | 0.114 | 1.568 | 3.213 | 2.342 | 2.236 | 1.795 | 289 | 488 | |
| Sonar-Deep-ResType=Web Agent, Timestamp=T12026.02 | 18.84 | 0.163 | 0.858 | 0.115 | 1.816 | 3.318 | 2.393 | 2.416 | 1.994 | 287 | 488 | |
| Gemini-2.5-FlashType=Web Agent, Timestamp=T22026.02 | 3.884 | 0.169 | 0.858 | 0.061 | 4.127 | 3.315 | 2.465 | 2.388 | 2.772 | 248 | 482 | |
| GPT-5-nanoType=GITSEARCH, Timestamp=T22026.02 | 3.26 | 0.145 | 0.848 | 0.043 | 4.266 | 3.668 | 2.656 | 2.734 | 3.141 | 245 | 488 | |
| Gemini-2.5-FlashType=Web Agent, Timestamp=T12026.02 | 3.213 | 0.171 | 0.857 | 0.043 | 4.242 | 3.334 | 2.465 | 2.418 | 2.809 | 255 | 488 | |
| GPT-5-nanoType=Web Agent, Timestamp=T12026.02 | 2.92 | 0.147 | 0.849 | 0.04 | 4.516 | 3.383 | 2.488 | 2.459 | 2.961 | 242 | 488 | |
| GPT-5-nanoType=Web Agent, Timestamp=T22026.02 | 2.846 | 0.145 | 0.849 | 0.039 | 4.496 | 3.34 | 2.465 | 2.426 | 2.984 | 242 | 488 | |
| Gemini-2.5-FlashType=GITSEARCH, Timestamp=T22026.02 | 2.238 | 0.178 | 0.861 | 0.047 | 4.477 | 3.744 | 2.682 | 2.744 | 3.114 | 210 | 484 | |
| GPT-5-nanoType=Supernote Lite2026.02 | 1.823 | 0.17 | 0.856 | 0.129 | 4.053 | 3.668 | 2.634 | 2.506 | 2.86 | 212 | 265 | |
| Llama-3.1-8BType=Supernote Lite2026.02 | 1.808 | 0.193 | 0.864 | 0.12 | 4.087 | 3.585 | 2.479 | 2.464 | 2.796 | 225 | 265 | |
| Ministral-8BType=Supernote Lite2026.02 | 1.8 | 0.208 | 0.861 | 0.125 | 4.083 | 3.589 | 2.487 | 2.355 | 2.672 | 154 | 265 | |
| Human-written Note2026.02 | 1.668 | — | — | — | — | — | — | — | — | 172 | 488 | |
| Apple-Nemotron-15BType=Supernote Lite2026.02 | 1.664 | 0.192 | 0.862 | 0.107 | 3.879 | 3.709 | 2.54 | 2.528 | 2.811 | 227 | 265 | |
| Qwen3-14BType=Supernote Lite2026.02 | 1.525 | 0.202 | 0.868 | 0.122 | 3.974 | 3.728 | 2.615 | 2.487 | 2.838 | 147 | 265 | |
| Gemini-2.5-FlashType=Supernote Lite2026.02 | 1.332 | 0.224 | 0.871 | 0.125 | 4.358 | 3.898 | 2.906 | 2.687 | 3.128 | 161 | 265 |