Intent Classification on POWERPLAY11 HINT3 (test)
76.1AccuracyDynamic Label Name Refinement
Evaluation Results
| Method | Links | |
|---|---|---|
| Dynamic Label Name RefinementBackbone=Llama3-8b-inst2024.12 | 76.1 | |
| Dynamic Label Name RefinementBackbone=Qwen2.5-7b-inst2024.12 | 74.76 | |
| RawBackbone=Qwen2.5-7b-inst2024.12 | 71.2 | |
| RawBackbone=Llama3-8b-inst2024.12 | 70.87 | |
| Chain-of-ThoughtBackbone=Qwen2.5-7b-inst2024.12 | 70.06 | |
| Chain-of-ThoughtBackbone=Llama3-8b-inst2024.12 | 67 | |
| Dynamic Label Name RefinementBackbone=Qwen2.5-1.5b-inst2024.12 | 63.1 | |
| RawBackbone=Qwen2.5-1.5b-inst2024.12 | 60.84 | |
| Chain-of-ThoughtBackbone=Qwen2.5-1.5b-inst2024.12 | 58.83 |