arXiv 论文：可解释性工具可跨模型复用，覆盖 110 组模型对验证
Interpretability Is Not Model-Specific
做模型可解释性的人值得看这篇：以前每个模型都要重新拟合拒绝方向和探针，现在证明换个模型也能直接用，加少量无标签 prompt 就能接近原生效果。覆盖五个模型家族、1B 到 72B，实验量够扎实。
一篇 arXiv 论文（2611.12675）研究了可解释性工具能否跨模型迁移。作者把拒绝方向、正确性方向、情感探针三类工具，在来自五个家族、参数量 1B 到 72B 的 110 组模型对之间迁移，同时覆盖 base 与 instruction-tuned 模型，并用五组随机种子做对照实验。
迁移效果取决于对齐映射所用的数据。仅在普通问题上拟合的无标签映射，能恢复目标模型自身拒绝调控效果的 54%，恢复正确性提升的 16% 到 77%；加入工具作用区域的无标签提示词后，两项指标分别提升到 102% 和 88% 到 98%，而对照方向的恢复率不超过 1%。论文还证明当两个模型在映射覆盖的数据上线性对应时，线性映射就会表现出这种特性，并据此提出一个无需标签的覆盖率评分，在超过 16,000 次运行中追踪恢复效果。
这项工作意味着可解释性结果一经发现即可跨模型复用，成本仅为少量无标签提示词，解释了此前跨域迁移失败的报告，也降低了对每个模型单独做可解释性分析的成本。
Interpretability Is Not Model-Specific
Interpretability tools are usually built for one model at a time: a refusal direction, a truthfulness probe, or a steering vector is found with labeled data inside the model it explains. We ask whether such tools can be carried to other models through a map between residual streams fit only on unlabeled prompts. We transfer a refusal direction, tested causally by addition and ablation, and two read-out tools, a correctness direction and a sentiment probe, across 110 pairs of models from five families and 1B to 72B parameters and between base and instruction-tuned models, with controls for specificity and five seeds. How much transfers depends on what the map was fit on. A map fit on ordinary questions recovers 54\% of the target's own refusal steering and 16\% to 77\% of its correctness gain; adding unlabeled prompts from the region where the tool operates raises these to 102\% and 88\% to 98\%, while control directions recover at most 1\%. We prove that a linear map behaves this way whenever two models correspond linearly on the data it covers, and a label-free coverage score derived from this result tracks recovery across more than 16,000 runs and explains earlier reports that maps between models fail across domains. Carried read-out tools outperform anything the target's unlabeled representations yield. Interpretability results, once found, can be reused across models at the cost of a few unlabeled prompts.
This report is categorized under AITOP's "Papers" desk, focusing on US frontier AI labs and Silicon Valley technology leaders. Within our cross-regional analytical framework, it directly maps to AI developer toolchains, coding agents, and engineering workflows. Key entities and technical topics under continuous tracking include 可解释性, arXiv, steering vector, 迁移学习.
Source attribution: curated from arXiv cs.AI (@Marcus Armstrong, Navid Ayoobi, Pradham Mummaleti, Alexander Chulzhanov, Arjun Mukherjee).
