J-Miner: Recovering Executable Decision Knowledge from Language-Model Classifiers
J-Miner:从语言模型分类器中恢复可执行的决策知识
机构 * Shanghai Research Institute for Intelligent Autonomous Systems, Tongji University(同济大学上海智能自主系统研究院) ; Shanghai Key Laboratory of Data Science, College of Computer Science and Artificial Intelligence, Fudan University(复旦大学计算机与人工智能学院上海市数据科学重点实验室) ; Meituan(美团) ; College of Design and Innovation, Tongji University(同济大学设计创意学院)
AI总结 J-Miner可从微调后的语言模型分类器中挖掘隐含的决策知识,生成可执行规则,规则复现源分类器决策精度高、保真度优,还能迁移至轻量级模型并保留高准确率。
Comments 19 pages, 12 figures, and 13 tables; includes appendices