Knowledge Editing on ZsRE (evaluation)
99ReliabilityONCEEDIT
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| ONCEEDITBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing, Plug-in model=Tiny-Llama2025.06 | 99 | 92 | 99 | 97 | 92 | 1 | |
| MEMITBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 98 | 92 | 76 | 89 | 78 | 3 | |
| ONCEEDITBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing, Plug-in model=Tiny-Llama2025.06 | 95 | 88 | 98 | 93 | 92 | 1 | |
| MEMITBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 91 | 89 | 50 | 77 | 78 | 3 | |
| WISEBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 84 | 78 | 99 | 87 | 82 | 3 | |
| ONCEEDITBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing, Plug-in model=Tiny-Llama2025.06 | 84 | 76 | 99 | 87 | 92 | 1 | |
| MEMITBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 78 | 76 | 53 | 69 | 78 | 3 | |
| WISEBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 76 | 68 | 100 | 81 | 82 | 3 | |
| WISEBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 68 | 64 | 99 | 77 | 82 | 3 | |
| DEFERBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 63 | 58 | 62 | 61 | 56 | 3 | |
| FT-LBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 55 | 41 | 54 | 50 | 32 | 3 | |
| DEFERBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 37 | 36 | 100 | 58 | 56 | 3 | |
| FT-LBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 34 | 21 | 13 | 23 | 32 | 3 | |
| DEFERBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 34 | 32 | 85 | 50 | 56 | 3 | |
| FT-LBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 11 | 10 | 48 | 23 | 32 | 3 | |
| ROMEBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 7 | 6 | 1 | 5 | 2 | 3 | |
| ROMEBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 1 | 1 | 1 | 1 | 2 | 3 | |
| ROMEBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 1 | 1 | 0 | 1 | 2 | 3 | |
| MENDBase Model=Llama2-7B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 0 | 0 | 0 | 0 | 0 | 3 | |
| MENDBase Model=Mistral-7B-v0.1, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 0 | 0 | 0 | 0 | 0 | 3 | |
| MENDBase Model=GPT-J-6B, Evaluation Setting=Teacher-forced, Editing Setting=Batch editing2025.06 | 0 | 0 | 0 | 0 | 0 | 3 |