Audio Hallucination Evaluation on Clotho-1K
16.98HRQwen2.5-Omni-7B
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Qwen2.5-Omni-7BNAICL=true2026.04 | 16.98 | 5.63 | 11.96 | 5.43 | 7.64 | |
| Gemini-2.5-pro2026.04 | 19.42 | 12.47 | 14.49 | 2.31 | 6.64 | |
| GPT-audio2026.04 | 21.81 | 15.28 | 15.28 | 3.02 | 7.64 | |
| Audio-Flamingo-32026.04 | 23.72 | 8.44 | 16.58 | 7.94 | 9.95 | |
| Qwen2-Audio-7B-Instruct2026.04 | 24.25 | 3.12 | 17.61 | 7.44 | 13.68 | |
| StepAudio22026.04 | 25.55 | 3.41 | 20.44 | 3.91 | 9.92 | |
| Qwen-audio-chat2026.04 | 25.93 | 5.23 | 20.3 | 3.62 | 10.55 | |
| MiMo-Audio-7B-Instruct2026.04 | 26.43 | 13.27 | 19.8 | 4.12 | 10.95 | |
| Qwen2.5-Omni-7B2026.04 | 26.53 | 7.24 | 18.19 | 7.84 | 10.25 | |
| Gemini-2.5-flash2026.04 | 26.85 | 13.78 | 20.26 | 4.15 | 9.93 | |
| Qwen2.5-Omni-3B2026.04 | 29.75 | 7.74 | 22.41 | 7.24 | 11.16 | |
| Kimi-Audio-7B-Instruct2026.04 | 36.48 | 7.34 | 28.74 | 7.44 | 19.3 | |
| GPT-audio-mini2026.04 | 37.69 | 23.92 | 25.53 | 4.02 | 14.59 | |
| SALMONN-7B2026.04 | 40.5 | 5.13 | 30.65 | 14.97 | 21.91 |