通过模块化技能包实现大语言模型的知识融合
Knowledge Fusion of Large Language Models Via Modular SkillPacks
AI总结：
GraftLLM通过模块化技能包实现大语言模型的知识融合，提升跨能力迁移的效率和效果。
摘要：
跨能力迁移是大语言模型（LLM）研究中的关键挑战，应用于多任务整合、模型压缩和持续学习。最近的工作如FuseLLM和FuseChat展示了将多种模型能力转移到轻量级模型的潜力，增强了适应性和效率，这激励我们研究更高效的跨能力迁移方法。然而，现有方法主要集中在小而同质的模型上，限制了其适用性。对于大而异质的模型，知识蒸馏与全参数微调往往忽视了学生模型的内在能力，并有灾难性遗忘的风险，而PEFT方法难以有效吸收源LLM的知识。为了解决这些问题，我们引入GraftLLM，一种新的方法，将源模型能力存储在目标模型中以SkillPack格式。该方法保留了通用能力，减少了参数冲突，并支持无遗忘的持续学习和模型融合。我们采用模块感知的自适应压缩策略来压缩参数更新，确保高效存储的同时保持任务特定的知识。所得到的SkillPack服务于一个紧凑且可转移的知识载体，非常适合异质模型融合和持续学习。在各种场景中的实验表明，GraftLLM在知识转移、知识融合和无遗忘学习方面优于现有技术，提供了一种可扩展且高效的跨能力迁移解决方案。代码在：https://github.com/duguodong7/GraftLLM上公开。
英文摘要：
Cross-capability transfer is a key challenge in large language model (LLM) research, with applications in multi-task integration, model compression, and continual learning. Recent works like FuseLLM and FuseChat have demonstrated the potential of transferring multiple model capabilities to lightweight models, enhancing adaptability and efficiency, which motivates our investigation into more efficient cross-capability transfer methods. However, existing approaches primarily focus on small, homogeneous models, limiting their applicability. For large, heterogeneous models, knowledge distillation with full-parameter fine-tuning often overlooks the student model's intrinsic capacity and risks catastrophic forgetting, while PEFT methods struggle to effectively absorb knowledge from source LLMs. To address these issues, we introduce GraftLLM, a novel method that stores source model capabilities in a target model with SkillPack format. This approach preserves general capabilities, reduces parameter conflicts, and supports forget-free continual learning and model fusion. We employ a module-aware adaptive compression strategy to compress parameter updates, ensuring efficient storage while maintaining task-specific knowledge. The resulting SkillPack serves as a compact and transferable knowledge carrier, ideal for heterogeneous model fusion and continual learning. Experiments across various scenarios demonstrate that GraftLLM outperforms existing techniques in knowledge transfer, knowledge fusion, and forget-free learning, providing a scalable and efficient solution for cross-capability transfer. The code is publicly available at: https://github.com/duguodong7/GraftLLM.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
