Intent Classification on Benchmark01 (test)
89.3In-Scope AccuracyGPT-3
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT-3LLM Intents=9, parameters=175B, Evaluation Mode=zero-shot2023.05 | 89.3 | 67 | |
| Flan-T5-XXLLLM Intents=5, parameters=11B, Evaluation Mode=zero-shot2023.05 | 86.5 | 43 | |
| Flan-T5-XXLLLM Intents=9, parameters=11B, Evaluation Mode=zero-shot2023.05 | 86.5 | 48 | |
| GPT-3LLM Intents=5, parameters=175B, Evaluation Mode=zero-shot2023.05 | 84.6 | 97 |