cs.CLSep 15, 2026

Cascade: Hierarchical Recoverability Control for Large Language Model Unlearning

Authors: Qingchen YuShiying DuanXiaodong LiYuhua WangZhiyu LiShiji ZhouYifan SunZhaoxin Fan

Organizations: 1Beijing Advanced Innovation Center for Future Blockchain and Privacy Computing, Beihang University · School of Artificial Intelligence, Beihang University · Center for Applied Statistics, School of Statistics, Renmin University of China · 4MemTensor (Shanghai) Technology Co., Ltd.

Abstract

Large Language Model (LLM) unlearning is essential for removing sensitive or copyrighted knowledge while preserving general utility. Existing methods often leave residual knowledge in intermediate representations, which can still be recovered. To address this, we propose Cascade, a hierarchical recoverability control framework that minimizes the internal identifiability of target knowledge. Cascade combines three complementary controls: path-level routing to suppress privacy-associated activation routes, representation-level compression to reduce geometric separability, and decoding-level intervention to limit residual recovery. Experiments on TOFU, MUSE-News, and WMDP, including robustness tests with query reformulation and extraction-style prompts, show that Cascade effectively reduces recoverability while maintaining stable model utility.

Explore similar work

CardsList