Code Generation on Truck CAN signal (Total)
98.04pass@1GPT-4o + DomRetriever
Evaluation results
| Method | Links | |
|---|---|---|
| GPT-4o + DomRetrieverBackbone=GPT-4o, Method Configuration=+DomRetriever (Qwen2.5-7B)2026.03 | 98.04 | |
| DomAgentBackbone=Qwen2.5-7B, Method Configuration=DomAgent2026.03 | 96.64 | |
| GPT-4oBackbone=GPT-4o, Method Configuration=Vanilla2026.03 | 71.22 | |
| Qwen2.5-7BBackbone=Qwen2.5-7B, Method Configuration=Vanilla2026.03 | 39.62 |