Mixed 20 Question on Mixed 20Q
87Acc (Animal)Fixed Round
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Fixed Roundtau=202026.01 | 87 | 78 | 20 | 10 | 1,440 | 600 | |
| Fixed Roundtau=152026.01 | 77 | 78 | 15 | 10 | 1,390 | 750 | |
| VOItau=0.012026.01 | 76 | 78 | 11.8 | 8.07 | 1,414 | 910 | |
| VOItau=0.052026.01 | 74 | 78 | 11.46 | 7.99 | 1,397 | 907 | |
| Adaptive2026.01 | 68 | 53 | 17.8 | 6.254 | 1,026 | 289 | |
| Fixed Roundtau=102026.01 | 60 | 78 | 10 | 10 | 1,270 | 830 | |
| Confidencetau=0.92026.01 | 59 | 65 | 8.48 | 6.49 | 1,149 | 784 | |
| Confidencetau=0.72026.01 | 45 | 60 | 5.68 | 4.56 | 989 | 743 | |
| Fixed RoundBackbone=Gemini-2.5-Flash, tau=152026.01 | 40 | 30 | 15 | 10 | 540 | -100 | |
| Fixed RoundBackbone=Gemini-2.5-Flash, tau=202026.01 | 39 | 30 | 20 | 10 | 480 | -360 | |
| ConfidenceBackbone=Gemini-2.5-Flash, tau=0.92026.01 | 36 | 30 | 11.28 | 9.25 | 538 | 50 | |
| Fixed RoundBackbone=Gemini-2.5-Flash, tau=102026.01 | 33 | 30 | 10 | 10 | 520 | 80 | |
| AdaptiveBackbone=Gemini-2.5-Flash2026.01 | 28 | 37 | 4.78 | 6.36 | 596 | 379 | |
| VOIBackbone=Gemini-2.5-Flash, tau=0.012026.01 | 28 | 55 | 8.48 | 7.63 | 738 | 368 | |
| Fixed Roundtau=52026.01 | 24 | 51 | 5 | 5 | 695 | 475 | |
| ConfidenceBackbone=Gemini-2.5-Flash, tau=0.52026.01 | 22 | 27 | 4.87 | 5.12 | 436 | 221 | |
| Confidencetau=0.52026.01 | 20 | 31 | 4.01 | 2.54 | 467 | 297 | |
| Fixed RoundBackbone=Gemini-2.5-Flash, tau=52026.01 | 16 | 29 | 5 | 5 | 395 | 175 | |
| ConfidenceBackbone=Gemini-2.5-Flash, tau=0.72026.01 | 16 | 31 | 5.06 | 6.08 | 413 | 187 | |
| VOIBackbone=Gemini-2.5-Flash, tau=0.052026.01 | 15 | 50 | 4.2 | 6.99 | 601 | 405 | |
| No QuestionBackbone=Gemini-2.5-Flash2026.01 | 1 | 6 | 0 | 0 | 70 | 70 | |
| No Question2026.01 | 1 | 6 | 0 | 0 | 70 | 70 |