Complex reasoning on VitaminC (test)
78.4Macro-F1EvoPool
Evaluation Results
| Method | Links | |
|---|---|---|
| EvoPoolBackbone=Llama-3.1-8B2026.06 | 78.4 | |
| LLM-AnnotationBackbone=Llama-3.1-8B2026.06 | 78.05 | |
| LLM-AnnotationBackbone=Qwen2-1.5B2026.06 | 74.11 | |
| EvoPoolBackbone=Qwen2-1.5B2026.06 | 73.83 | |
| AlchemistBackbone=Qwen2-1.5B2026.06 | 72.67 | |
| LLM-AnnotationBackbone=RoBERTa-large2026.06 | 70.97 | |
| EvoPoolBackbone=RoBERTa-large2026.06 | 70.38 | |
| EvoPoolDownstream Model=Llama-3.1-8B2026.06 | 67.97 | |
| EvoPoolBackbone=gpt-4o-mini2026.06 | 66.67 | |
| EvoPoolDownstream Model=Qwen3-1.7B2026.06 | 66.59 | |
| LLM-AnnotationDownstream Model=Qwen3-1.7B2026.06 | 65.62 | |
| LLM-AnnotationDownstream Model=Llama-3.1-8B2026.06 | 65.08 | |
| LLM annotationBackbone=gpt-4o-mini2026.06 | 64.82 | |
| LLM-EvalDownstream Model=RoBERTa-large2026.06 | 64.82 | |
| LLM-EvalDownstream Model=Qwen3-1.7B2026.06 | 64.82 | |
| LLM-EvalDownstream Model=Llama-3.1-8B2026.06 | 64.82 | |
| EvoPoolDownstream Model=RoBERTa-large2026.06 | 64.76 | |
| LLM-AnnotationDownstream Model=RoBERTa-large2026.06 | 59.19 | |
| DataSculptBackbone=Qwen2-1.5B2026.06 | 53.66 | |
| Self-TrainDownstream Model=RoBERTa-large2026.06 | 52.82 | |
| Self-TrainDownstream Model=Qwen3-1.7B2026.06 | 51.74 | |
| GoldenDownstream Model=Qwen3-1.7B2026.06 | 50.65 | |
| Self-TrainDownstream Model=Llama-3.1-8B2026.06 | 47.84 | |
| AlchemistBackbone=Llama-3.1-8B2026.06 | 46.42 | |
| GoldenDownstream Model=RoBERTa-large2026.06 | 45.43 | |
| GoldenDownstream Model=Llama-3.1-8B2026.06 | 44.81 | |
| DataSculptBackbone=Llama-3.1-8B2026.06 | 43.19 | |
| DataSculptBackbone=RoBERTa-large2026.06 | 39.97 | |
| DataSculptDownstream Model=RoBERTa-large2026.06 | 26.46 | |
| DataSculptDownstream Model=Llama-3.1-8B2026.06 | 26 | |
| DataSculptDownstream Model=Qwen3-1.7B2026.06 | 22.57 | |
| AlchemistBackbone=gpt-4o-mini2026.06 | 16.81 | |
| AlchemistBackbone=RoBERTa-large2026.06 | 16.67 | |
| AlchemistDownstream Model=RoBERTa-large2026.06 | 16.67 | |
| AlchemistDownstream Model=Qwen3-1.7B2026.06 | 16.67 | |
| AlchemistDownstream Model=Llama-3.1-8B2026.06 | 16.67 | |
| DataSculptBackbone=gpt-4o-mini2026.06 | 4.73 |