Competition Math Accuracy on HMMT Nov 25
59.4AccuracySELFCOMPACT
Evaluation results
| Method | Links | |
|---|---|---|
| SELFCOMPACTModel=Qwen3-30B-A3B-Instruct-2507, Avg. tokens per question=29k, Thinking state=Enabled2026.06 | 59.4 | |
| Fixed Interval SummaryModel=Qwen3-30B-A3B-Instruct-2507, Avg. tokens per question=26k, Thinking state=Enabled2026.06 | 57.3 | |
| No CompactionModel=Qwen3-30B-A3B-Instruct-2507, Avg. tokens per question=16k, Thinking state=Enabled2026.06 | 54.7 | |
| SELFCOMPACTModel=Qwen3.5-9B, Avg. tokens per question=93k, Thinking state=Disabled2026.06 | 48.2 | |
| SELFCOMPACTModel=Qwen3-4B-Instruct-2507, Avg. tokens per question=48k, Thinking state=Enabled2026.06 | 47.8 | |
| Fixed Interval SummaryModel=Qwen3-4B-Instruct-2507, Avg. tokens per question=44k, Thinking state=Enabled2026.06 | 44 | |
| Fixed Interval SummaryModel=Qwen3.5-9B, Avg. tokens per question=90k, Thinking state=Disabled2026.06 | 42.1 | |
| No CompactionModel=Qwen3-4B-Instruct-2507, Avg. tokens per question=16k, Thinking state=Enabled2026.06 | 40.8 | |
| No CompactionModel=Qwen3.5-9B, Avg. tokens per question=16k, Thinking state=Disabled2026.06 | 38.2 | |
| SELFCOMPACTModel=Qwen3.5-4B, Avg. tokens per question=67k, Thinking state=Disabled2026.06 | 37.4 | |
| Fixed Interval SummaryModel=Qwen3.5-4B, Avg. tokens per question=64k, Thinking state=Disabled2026.06 | 35.5 | |
| No CompactionModel=Qwen3.5-4B, Avg. tokens per question=16k, Thinking state=Disabled2026.06 | 26.4 |