让大模型轻松新增领域能力，只需改一个模块。
SemiAdapt-Instruct: Extensible Instruction Tuning via Latent Domain-Specialised Adapters
- 通过发现隐含指令领域，为每个领域训练独立LoRA适配器。
- 新领域仅需更新单个适配器，性能超越全模型微调。
- 适合需要持续扩展能力的生产级大模型系统使用。
指令微调的大语言模型部署在不断演化的领域环境中，但不进行完整重训练即可扩展模型能力仍是未解难题。我们提出SemiAdapt-Instruct，一种模块化框架：自动发现潜在指令领域，并行训练各领域LoRA适配器，实现无参数路由。新增领域仅需单适配器训练，无需修改已有组件。在ROUGE-L和LLM作为裁判的评估中，SemiAdapt-Instruct在所有配置下均优于全模型微调，性能等同于单个LoRA微调，并提供单体方法无法实现的可扩展性。实证表明，仅用新领域数据更新一个适配器，即可超越所有单体基线。研究还发现，不同独立发现方法收敛到相同可特化的领域。这些结果证明，将异构指令数据分解为潜在领域，可构建可扩展的NLP系统，演化领域仅需针对性地更新单个适配器，无需全模型重训。
原文摘要 · Abstract (English)
Instruction-tuned LLMs are deployed into environments where domains evolve, yet extending a fine-tuned model's capabilities without full retraining remains an unsolved practical challenge. We present SemiAdapt-Instruct, a modular framework that discovers latent instruction domains, trains per-domain LoRA adapters in parallel, and performs parameter-free routing, incorporating new domains via single-adapter training without modifying existing components. SemiAdapt-Instruct outperforms full model fine-tuning across all configurations on both ROUGE-L and LLM-as-a-judge evaluation, while matching single LoRA fine-tuning and delivering extensibility that monolithic approaches cannot provide. We empirically demonstrate this extensibility by showing that updating a single adapter with new domain data outperforms all monolithic baselines. Our study also finds that independent discovery methods converge on the same specialisation-friendly domains. These findings demonstrate that decomposing heterogeneous instruction data into latent domains enables extensible NLP systems where evolving domains require only targeted single-adapter updates, eliminating the need for full model retraining.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。