Spoken Question Answering on MCIF long track
40.86BERTScoreN2+Avg
Evaluation results
| Method | Links | |
|---|---|---|
| N2+AvgPrompt Evaluation Mode=Mixed2026.06 | 40.86 | |
| N2+AvgPrompt Evaluation Mode=Fixed2026.06 | 40.68 | |
| N2+IT+AvgPrompt Evaluation Mode=Fixed2026.06 | 40.42 | |
| N2+IT+AvgPrompt Evaluation Mode=Mixed2026.06 | 40.31 | |
| N2Prompt Evaluation Mode=Fixed2026.06 | 39.98 | |
| N2Prompt Evaluation Mode=Mixed2026.06 | 39.98 | |
| Temp. Sampling (T = 2)Prompt Evaluation Mode=Mixed2026.06 | 37.87 | |
| Temp. Sampling (T = 2)Prompt Evaluation Mode=Fixed2026.06 | 37.75 | |
| Fixed-Prob. SamplingPrompt Evaluation Mode=Mixed2026.06 | 36.42 | |
| Fixed-Prob. SamplingPrompt Evaluation Mode=Fixed2026.06 | 36.16 | |
| Chain-of-Thought (CoT)Prompt Evaluation Mode=Mixed2026.06 | 35.5 | |
| Chain-of-Thought (CoT)Prompt Evaluation Mode=Fixed2026.06 | 34.88 | |
| Cascaded+FT (Contrastive)Prompt Evaluation Mode=Fixed2026.06 | 33.36 | |
| Qwen 2.5 OmniPrompt Evaluation Mode=Mixed2026.06 | 32.94 | |
| Cascaded+FT (Contrastive)Prompt Evaluation Mode=Mixed2026.06 | 32.09 | |
| Qwen 2.5 OmniPrompt Evaluation Mode=Fixed2026.06 | 30.78 | |
| CascadedPrompt Evaluation Mode=Mixed2026.06 | 27.66 | |
| CascadedPrompt Evaluation Mode=Fixed2026.06 | 27.48 | |
| N2+IT+SHASPrompt Evaluation Mode=Fixed2026.06 | -11.09 | |
| N2+IT+SHASPrompt Evaluation Mode=Mixed2026.06 | -12.14 |