Question Answering on IndiRef Spot the difference
42Temporal Accuracy (FEM)Qwen QwQ - 32B
Evaluation Results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Qwen QwQ - 32BContext=Full Dialog History2026.01 | 42 | 34 | 26 | 22 | 26 | 20 | 28 | 20 | |
| Llama3.1 - 8BContext=Full Dialog History2026.01 | 40 | 38 | 30 | 32 | 36 | 36 | 26 | 16 | |
| Gemma2 - 27BContext=Full Dialog History2026.01 | 40 | 30 | 28 | 20 | 32 | 26 | 26 | 12 | |
| Gemma2 - 9BContext=Full Dialog History2026.01 | 34 | 20 | 24 | 16 | 28 | 16 | 20 | 10 | |
| Gemma2 - 2BContext=Full Dialog History2026.01 | 18 | 12 | 18 | 10 | 20 | 12 | 16 | 10 |