Length Prediction on ForeLen Reasoning 1.0 (test)
138.01MAETRAIL
Evaluation Results
| Method | Links | |
|---|---|---|
| TRAILModel=Qwen2.5 1.5B2026.02 | 138.01 | |
| LTR-CModel=Qwen2.5 1.5B2026.02 | 139.02 | |
| EGTPModel=Qwen2.5 0.5B2026.02 | 145.85 | |
| TRAILModel=Qwen2.5 0.5B2026.02 | 146.24 | |
| EGTPModel=Qwen2.5 1.5B2026.02 | 146.48 | |
| S3Model=Qwen2.5 1.5B2026.02 | 169.72 | |
| LTR-CModel=Qwen2.5 0.5B2026.02 | 178.28 | |
| S3Model=Qwen2.5 0.5B2026.02 | 200.91 | |
| SSJF-RegModel=Qwen2.5 0.5B2026.02 | 221.15 | |
| PiAModel=Qwen2.5 0.5B2026.02 | 296.02 | |
| SSJF-RegModel=Qwen2.5 1.5B2026.02 | 310.64 | |
| PiAModel=Qwen2.5 1.5B2026.02 | 432.05 | |
| SSJF-MCModel=Qwen2.5 0.5B2026.02 | 466 | |
| TPVModel=Qwen2.5 1.5B2026.02 | 584.08 | |
| TPVModel=Qwen2.5 0.5B2026.02 | 597 | |
| SSJF-MCModel=Qwen2.5 1.5B2026.02 | 815.78 |