PDF Exposure Analysis on Corpus B one-page corpus
0Exposed Gap CountGemini 3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Gemini 3Interface=Web Chat2026.06 | 0 | 12 | |
| Gemini 3Interface=Official API2026.06 | 12 | — | |
| Qwen 3.5Interface=Web Chat2026.06 | 13 | 4 | |
| Kimi K2.6Interface=Web Chat2026.06 | 14 | 3 | |
| GLM-5.1Interface=Official API2026.06 | 14 | — | |
| GLM-5.1Interface=Web Chat2026.06 | 14 | 0 | |
| Sonnet 4.6Interface=Official API2026.06 | 15 | — | |
| Grok 4.2Interface=Official API2026.06 | 17 | — | |
| Kimi K2.6Interface=Official API2026.06 | 17 | — | |
| Qwen-LongInterface=Official API2026.06 | 17 | — | |
| Grok 4.2Interface=Web Chat2026.06 | 18 | 1 | |
| Sonnet 4.6Interface=Web Chat2026.06 | 19 | 4 | |
| GPT-5.4Interface=Web Chat2026.06 | 19 | 2 | |
| GPT-5.4Interface=Official API2026.06 | 21 | — |