SyncAI.news, a Varaisys broadcasting
J-Miner: Recovering the Decision Logic of Fine-Tuned LLM Classifiers as Compact Rules
YG

Yunfan Gao, Xinyi Huang, Tao Sheng, Haorui Song, Yun Xiong, Haofen Wang

· 1 min read

ResearcharXiv cs.CL

J-Miner: Recovering the Decision Logic of Fine-Tuned LLM Classifiers as Compact Rules

arXiv:2608.17063v2 Announce Type: replace-cross Abstract: Task-fine-tuned large language model (LLM) classifiers acquire task-specific decision knowledge, but this knowledge remains implicit in distributed internal computations, making their decision logic difficult to interpret. We introduce the Executable Decision Compression (EDC) framework and propose J-Miner, which mines vocabulary-named variables from internal readouts and learns rules shared across inputs to produce executable explanations. Analysis reveals that a small set of these variables captures much of the classifier's decision behavior, holding for both varying parameter scales within a family and distinct families. Across six binary tasks, a rule using just one variable reproduces 76.7% of source-classifier decisions on average, rising to 88.8% with 16 variables. Most of the decision information retained by these variables comes from internal activations beyond literal surface matching. A lightweight text reader predicts the variable states, allowing the same fixed rules to execute independently of the source classifier.

Original source

This story was published by arXiv cs.CL and written by Yunfan Gao, Xinyi Huang, Tao Sheng, Haorui Song, Yun Xiong, Haofen Wang. SyncAI.news shows a preview; the complete article is on the publisher's site.

Read the full story on arxiv.org

Similar News