Feasibility Classification on Real-world queries (N = 1,710)
96AccuracyGemini 3.0 Pro
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| Gemini 3.0 Pro2026.04 | 96 | 95 | 100 | 97 | |
| Gemini 3.0 Flash2026.04 | 95 | 93 | 100 | 96 | |
| GPT-5.22026.04 | 92 | 90 | 100 | 94 | |
| GPT-5-Mini2026.04 | 91 | 90 | 97 | 94 | |
| Llama-4-17B2026.04 | 86 | 87 | 95 | 91 |