Reasoning on NumS (test)
100AccuracyTCR-gold
Evaluation Results
| Method | Links | |
|---|---|---|
| TCR-goldBackbone=Qwen3-8B-Instruct, Intervention Method=TCR-gold2026.01 | 100 | |
| TCRBackbone=Qwen3-8B-Instruct, Intervention Method=TCR2026.01 | 99 | |
| Qwen3-8B-InstructBackbone=Qwen3-8B-Instruct, Intervention Method=None2026.01 | 98 | |
| DoLaBackbone=Qwen3-8B-Instruct, Intervention Method=DoLa2026.01 | 98 | |
| TCR-goldBackbone=Phi-3-Instruct, Intervention Method=TCR-gold2026.01 | 82 | |
| TCRBackbone=Phi-3-Instruct, Intervention Method=TCR2026.01 | 76 | |
| Phi-3-InstructBackbone=Phi-3-Instruct, Intervention Method=None2026.01 | 72 | |
| DoLaBackbone=Phi-3-Instruct, Intervention Method=DoLa2026.01 | 72 | |
| TCR-goldBackbone=Qwen2.5-7B-Instruct, Intervention Method=TCR-gold2026.01 | 54.5 | |
| DoLaBackbone=Qwen2.5-7B-Instruct, Intervention Method=DoLa2026.01 | 48.7 | |
| TCRBackbone=Qwen2.5-7B-Instruct, Intervention Method=TCR2026.01 | 46 | |
| Qwen2.5-7B-InstructBackbone=Qwen2.5-7B-Instruct, Intervention Method=None2026.01 | 41.1 | |
| TCR-goldBackbone=LLaMA3-8B-Instruct, Intervention Method=TCR-gold2026.01 | 10.1 | |
| TCRBackbone=LLaMA3-8B-Instruct, Intervention Method=TCR2026.01 | 7.8 | |
| DoLaBackbone=LLaMA3-8B-Instruct, Intervention Method=DoLa2026.01 | 6.9 | |
| LLaMA3-8B-InstructBackbone=LLaMA3-8B-Instruct, Intervention Method=None2026.01 | 4.5 |