Safety Evaluation on TrustLLM
8.26ASRChat model
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| Chat modelBackbone LLM=Qwen-2.5 7B2026.01 | 8.26 | 65.81 | — | |
| Chat modelBackbone LLM=LLaMa-2 7B2026.01 | 12.86 | 66.55 | — | |
| Chat modelBackbone LLM=LLaMa-3.1 8B2026.01 | 13.88 | 62.78 | — | |
| ITIBackbone LLM=Qwen-2.5 7B2026.01 | 14.99 | 59.83 | — | |
| ARREST-SBBackbone LLM=Qwen-2.5 7B2026.01 | 18.82 | 69.36 | — | |
| ARREST-SCBackbone LLM=Qwen-2.5 7B2026.01 | 22.75 | 70.19 | — | |
| Chat modelBackbone LLM=Yi-1.5 9B2026.01 | 23.08 | 58.69 | — | |
| ARREST-SBBackbone LLM=LLaMa-3.1 8B2026.01 | 29.98 | 70 | — | |
| ITIBackbone LLM=LLaMa-3.1 8B2026.01 | 30.41 | 67.97 | — | |
| Base ModelBackbone LLM=Qwen-2.5 7B2026.01 | 34.99 | 59.2 | — | |
| ARREST-SCBackbone LLM=LLaMa-3.1 8B2026.01 | 36.37 | 79.25 | — | |
| ARREST-SCBackbone LLM=LLaMa-2 7B2026.01 | 37.39 | 75.3 | — | |
| ARREST-SCBackbone LLM=Yi-1.5 9B2026.01 | 45.82 | 76.01 | — | |
| ARREST-SBBackbone LLM=Yi-1.5 9B2026.01 | 47.87 | 67.38 | — | |
| ITIBackbone LLM=Yi-1.5 9B2026.01 | 49.82 | 66.87 | — | |
| Base ModelBackbone LLM=LLaMa-3.1 8B2026.01 | 52.03 | 65.5 | — | |
| ARREST-SBBackbone LLM=LLaMa-2 7B2026.01 | 52.55 | 75.04 | — | |
| ITIBackbone LLM=LLaMa-2 7B2026.01 | 56.98 | 71.38 | — | |
| Base ModelBackbone LLM=Yi-1.5 9B2026.01 | 66.51 | 47.55 | — | |
| Base ModelBackbone LLM=LLaMa-2 7B2026.01 | 70.35 | 48.11 | — | |
| IQuest-Coder-V1-40B-InstructParameters=40B, Type=Instruct2026.03 | — | — | 65 | |
| IQuest-Coder-V1-40B-ThinkingParameters=40B, Type=Thinking2026.03 | — | — | 86.4 | |
| Qwen2.5-Coder-32B-InstructParameters=32B, Type=Instruct2026.03 | — | — | 70.2 | |
| Qwen3-Coder-480B-A35B-InstructParameters=480B-A35B, Type=Instruct2026.03 | — | — | 88.4 |