Utterance Classification on Key shifts 4 turns Phase 2
83F1 ScoreClaude Sonnet
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Claude SonnetPrompt Condition=State-aug.2026.05 | 83 | 30 | |
| Claude SonnetPrompt Condition=Definitions2026.05 | 74 | 25 | |
| GPT-4oPrompt Condition=State-aug.2026.05 | 74 | 50 | |
| Llama-3.1-8B-InstructPrompt Condition=State-aug.2026.05 | 56 | 15 | |
| Claude SonnetPrompt Condition=Minimal2026.05 | 50 | 5 | |
| GPT-4oPrompt Condition=Definitions2026.05 | 50 | 10 | |
| Qwen2.5-7B-InstructPrompt Condition=Definitions2026.05 | 47 | 0 | |
| Qwen2.5-7B-InstructPrompt Condition=State-aug.2026.05 | 47 | 0 | |
| Llama-3.1-8B-InstructPrompt Condition=Definitions2026.05 | 47 | 20 | |
| GPT-4oPrompt Condition=Minimal2026.05 | 43 | 0 | |
| Qwen2.5-7B-InstructPrompt Condition=Minimal2026.05 | 39 | 10 | |
| Llama-3.1-8B-InstructPrompt Condition=Minimal2026.05 | 37 | 0 |