| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Long-form Generation | LongGenBench | CR80.03 | 24 | |
| Long-context reasoning | LongGenBench 8K | GSM8K Score44.51 | 22 | |
| Long-context reasoning | LongGenBench 4K | GSM8K Score53.18 | 22 | |
| Long-context Question Answering | LONGGENBENCH n=30 | CSQA74.1 | 5 | |
| Long Text Generation | LongGenBench 32K | CR84.95 | 4 | |
| Long Text Generation | LongGenBench 16K | CR98.51 | 4 | |
| Long-context generation | LongGenBench | Completion Rate97.627 | 3 |