让不同语言模型共享激活接口，工具无需重做即可复用。
One Adapter Pair per Model: A Universal Activation Interface for Language Models
- 用轻量适配器建立统一激活空间，跨模型共用
- 5个模型间语义相近文本在共享空间中聚类一致
- 适合需要跨模型工具复用的研究者使用
基于激活的工具通常依赖于特定模型的原始隐藏空间，需为每个新语言模型重新构建探测器、稀疏自编码器和自然语言解释器。我们提出通用激活总线（Universal Activation Bus）框架，实现兼容语言模型间的统一激活接口。利用少量源模型，学习一个共享的密集空间及每个模型对应的轻量级线性编码-解码适配器对。源模型训练完成后接口冻结；新模型仅需在未标注的对应文本上拟合其适配器即可接入。该接口使激活工具可跨连接模型共享，包括通用探测器、SAE特征以及原为其他模型训练的自然语言解释器。在五个模型上，语义相关的文本在共享空间中形成一致邻域，接入模型能有效复用这些工具而无需重新训练。进一步证明，一个模型的中间激活可被另一模型的冻结上层直接使用以生成预测。结果建立了稳定、模型级别的激活契约，支持兼容模型间可复用工具。
原文摘要 · Abstract (English)
Activation-based tools are usually tied to one model's native hidden space, requiring probes, sparse autoencoders, and natural-language interpreters to be rebuilt or rediscovered for each new language model. We present a Universal Activation Bus, a framework that provides a common activation interface across compatible language models. Using a small set of source models, we learn a shared dense space together with one lightweight linear encoder--decoder adapter pair per model. After source training, the interface is frozen; a new model joins by fitting only its adapter pair on unlabeled matched text. The resulting interface allows activation-based tools to be shared across connected models, including common probes and SAE features as well as access to an NLA originally trained for a different model. Across five models, semantically related texts form consistent neighborhoods in the shared space, and an onboarded model reuses these tools effectively without retraining them. We further show that an intermediate activation from one model can be used by another model's frozen upper layers to produce predictions. These results establish a stable, model-wise activation contract for reusable tools across compatible language models.
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。