Do Not Merge My Model! Safeguarding Open-Source LLMs Against Unauthorized Model Merging
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915628049235968 |
|---|---|
| author | Li, Qinfeng Pan, Miao Chen, Jintao Teng, Fu Shen, Zhiqiang Su, Ge Peng, Hao Zhang, Xuhong |
| author_facet | Li, Qinfeng Pan, Miao Chen, Jintao Teng, Fu Shen, Zhiqiang Su, Ge Peng, Hao Zhang, Xuhong |
| contents | Model merging has emerged as an efficient technique for expanding large language models (LLMs) by integrating specialized expert models. However, it also introduces a new threat: model merging stealing, where free-riders exploit models through unauthorized model merging. Unfortunately, existing defense mechanisms fail to provide effective protection. Specifically, we identify three critical protection properties that existing methods fail to simultaneously satisfy: (1) proactively preventing unauthorized merging; (2) ensuring compatibility with general open-source settings; (3) achieving high security with negligible performance loss. To address the above issues, we propose MergeBarrier, a plug-and-play defense that proactively prevents unauthorized merging. The core design of MergeBarrier is to disrupt the Linear Mode Connectivity (LMC) between the protected model and its homologous counterparts, thereby eliminating the low-loss path required for effective model merging. Extensive experiments show that MergeBarrier effectively prevents model merging stealing with negligible accuracy loss. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_10712 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Do Not Merge My Model! Safeguarding Open-Source LLMs Against Unauthorized Model Merging Li, Qinfeng Pan, Miao Chen, Jintao Teng, Fu Shen, Zhiqiang Su, Ge Peng, Hao Zhang, Xuhong Cryptography and Security Artificial Intelligence Model merging has emerged as an efficient technique for expanding large language models (LLMs) by integrating specialized expert models. However, it also introduces a new threat: model merging stealing, where free-riders exploit models through unauthorized model merging. Unfortunately, existing defense mechanisms fail to provide effective protection. Specifically, we identify three critical protection properties that existing methods fail to simultaneously satisfy: (1) proactively preventing unauthorized merging; (2) ensuring compatibility with general open-source settings; (3) achieving high security with negligible performance loss. To address the above issues, we propose MergeBarrier, a plug-and-play defense that proactively prevents unauthorized merging. The core design of MergeBarrier is to disrupt the Linear Mode Connectivity (LMC) between the protected model and its homologous counterparts, thereby eliminating the low-loss path required for effective model merging. Extensive experiments show that MergeBarrier effectively prevents model merging stealing with negligible accuracy loss. |
| title | Do Not Merge My Model! Safeguarding Open-Source LLMs Against Unauthorized Model Merging |
| topic | Cryptography and Security Artificial Intelligence |
| url | https://arxiv.org/abs/2511.10712 |