用文本生成适配器，让大模型按特定文化或立场输出，参数少26倍
Zhyper: Factorized Hypernetworks for Conditioned LLM Fine-Tuning
- 通过文本描述生成上下文感知的LoRA适配器，实现高效微调
- 在多个评测中性能接近顶尖方法，参数量减少最多26倍
- 适合需要低成本定制大模型行为的研究者和应用开发者
大型语言模型的条件化是指指导其生成符合特定文化规范、政治立场或任意指定语义要求的内容。然而，提示工程无法保证模型行为与目标条件一致，因预训练和对齐数据集存在归纳偏置。以往工作通过直接调节LoRA权重进行微调，但引入大量参数。为此，我们提出Zhyper，一种参数高效的分解式超网络框架，可从文本描述生成上下文感知的LoRA适配器。多基准测试表明，Zhyper性能媲美最先进基线，参数量最多减少26倍。此外，我们将Zhyper扩展至文化对齐任务，展现出对域外场景更好的泛化能力及对细微上下文价值更优的捕捉效果。
原文摘要 · Abstract (English)
Large Language Model (LLM) conditioning refers to instructing an LLM to generate content in accordance with the norms and values of a specific culture, beliefs of a particular political orientation, or any desired text-specified semantic conditioning. Unfortunately, prompt engineering does not ensure that LLMs behave in accordance with a desired conditioning due to the inductive bias of the pre-training and alignment datasets. Prior works have focused on fine-tuning LLMs by directly conditioning the LoRA weights; however, such methods introduce a large number of parameters. As a remedy, we propose Zhyper, a parameter-efficient factorized hypernetwork framework that generates context-aware LoRA adapters from textual descriptions. Experiments on multiple benchmarks show that Zhyper achieves competitive performance with up to 26x fewer parameters than the state-of-the-art baselines. Furthermore, we extend Zhyper to cultural alignment, demonstrating improved generalization to out-of-domain settings and a better capturing of fine-grained contextual values.
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。