Reading Comprehension on KnowsRecord v1.0 (SHORT (<2K words))
61Accuracy (PDF)Strong Tier
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Strong TierModel Tier=Strong, Models=MiMO-V2-Pro, Kimi-K2.52026.04 | 61 | 14 | |
| Weak TierModel Tier=Weak, Models=Qwen3.5-0.8B, Qwen3.5-2B2026.04 | 58 | 8 | |
| Medium TierModel Tier=Medium, Models=MiMO-V2-Flash, Qwen3.5-27B2026.04 | 50 | 8 |