Knowledge Editing on MQuAKE (Edit Success Rate and Neighborhood Score)
99.95Edit Success RateCoT2Edit
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| CoT2EditModel=LLaMA3-8B2026.04 | 99.95 | 92.04 | — | — | |
| CoT2EditModel=Falcon3-10B2026.04 | 99.48 | 92.69 | — | — | |
| CoT2EditModel=Qwen3-14B2026.04 | 99.22 | 92.24 | — | — | |
| AlphaEditModel=Falcon3-10B2026.04 | 97.73 | 71.69 | — | — | |
| LTEModel=Qwen3-14B2026.04 | 97.73 | 77.65 | — | — | |
| AlphaEditModel=Qwen3-14B2026.04 | 97.71 | 76.67 | — | — | |
| AlphaEditModel=LLaMA3-8B2026.04 | 97.11 | 73.63 | — | — | |
| EditCoTModel=Falcon3-10B2026.04 | 96.73 | 81.11 | — | — | |
| GLAMEModel=Falcon3-10B2026.04 | 96.68 | 82.31 | — | — | |
| PMETModel=LLaMA3-8B2026.04 | 96.37 | 70.28 | — | — | |
| EditCoTModel=Qwen3-14B2026.04 | 96.37 | 83.82 | — | — | |
| GLAMEModel=Qwen3-14B2026.04 | 95.56 | 84.31 | — | — | |
| LTEModel=Falcon3-10B2026.04 | 94.51 | 79.18 | — | — | |
| PMETModel=Qwen3-14B2026.04 | 94.13 | 79.78 | — | — | |
| LTEModel=LLaMA3-8B2026.04 | 93.46 | 74.45 | — | — | |
| GLAMEModel=LLaMA3-8B2026.04 | 93.28 | 77.82 | — | — | |
| PMETModel=Falcon3-10B2026.04 | 93.16 | 77.48 | — | — | |
| MEMITModel=Qwen3-14B2026.04 | 91.65 | 73.48 | — | — | |
| SKEMEModel=Falcon3-10B2026.04 | 90.98 | 67.43 | — | — | |
| EditCoTModel=LLaMA3-8B2026.04 | 90.58 | 78.71 | — | — | |
| SKEMEModel=LLaMA3-8B2026.04 | 88.05 | 67.12 | — | — | |
| MEMITModel=Falcon3-10B2026.04 | 87.68 | 70.34 | — | — | |
| IKEModel=Falcon3-10B2026.04 | 87.44 | 64.29 | — | — | |
| SKEMEModel=Qwen3-14B2026.04 | 87.28 | 66.13 | — | — | |
| IKEModel=LLaMA3-8B2026.04 | 86.63 | 65.38 | — | — | |
| IKEModel=Qwen3-14B2026.04 | 84.93 | 64.29 | — | — | |
| MEMITModel=LLaMA3-8B2026.04 | 81.19 | 64.03 | — | — | |
| RoMEModel=Qwen3-14B2026.04 | 80.74 | 70.15 | — | — | |
| RoMEModel=LLaMA3-8B2026.04 | 80.28 | 60.38 | — | — | |
| RoMEModel=Falcon3-10B2026.04 | 80.13 | 65.21 | — | — | |
| BaseModel=Qwen3-1.7B2026.05 | — | — | 10 | 100 | |
| BaseModel=Qwen3-4B-It2026.05 | — | — | 16 | 94 | |
| BaseModel=Qwen3-8B2026.05 | — | — | 19 | 94 | |
| MEMITModel=Qwen3-1.7B2026.05 | — | — | 60 | 98 | |
| MEMITModel=Qwen3-4B-It2026.05 | — | — | 53 | 90 | |
| MEMITModel=Qwen3-8B2026.05 | — | — | 70 | 89 | |
| MIXSDModel=Qwen3-1.7B, lambda=02026.05 | — | — | 78 | 67 | |
| MIXSDModel=Qwen3-1.7B, lambda=0.32026.05 | — | — | 97 | 84 | |
| MIXSDModel=Qwen3-1.7B, lambda=0.52026.05 | — | — | 95 | 76 | |
| MIXSDModel=Qwen3-4B-It, lambda=02026.05 | — | — | 81 | 92 | |
| MIXSDModel=Qwen3-4B-It, lambda=0.32026.05 | — | — | 97 | 95 | |
| MIXSDModel=Qwen3-4B-It, lambda=0.52026.05 | — | — | 69 | 88 | |
| MIXSDModel=Qwen3-8B, lambda=02026.05 | — | — | 78 | 73 | |
| MIXSDModel=Qwen3-8B, lambda=0.32026.05 | — | — | 99 | 86 | |
| MIXSDModel=Qwen3-8B, lambda=0.52026.05 | — | — | 90 | 86 | |
| SFTModel=Qwen3-1.7B2026.05 | — | — | 100 | 53 | |
| SFTModel=Qwen3-4B-It2026.05 | — | — | 100 | 41 | |
| SFTModel=Qwen3-8B2026.05 | — | — | 100 | 58 |