DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillation

v1v2 (latest)

DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillation

26 May 2025

ArXiv (abs)PDF HTML Github (3★)

Papers citing "DOGe: Defensive Output Generation for LLM Protection Against Knowledge Distillation"

4 / 4 papers shown

Title
Leave It to the Experts: Detecting Knowledge Distillation via MoE Expert Signatures Pingzhi Li Morris Yu-Chao Huang Zhen Tan Qingquan Song Jie Peng Kai Zou Yu Cheng Kaidi Xu Tianlong Chen MoE AAML 233 0 0 19 Oct 2025
Information-Preserving Reformulation of Reasoning Traces for Antidistillation Jiayu Ding Lei Cui Li Dong Nanning Zheng Furu Wei LRM 120 0 0 13 Oct 2025
Unified attacks to large language model watermarks: spoofing and scrubbing in unauthorized knowledge distillationKnowledge-Based Systems (KBS), 2025 Xin Yi Shunfan Zhengc Linlin Wanga Xiaoling Wang Xiaoling Wang Liang He AAML 1.1K 3 0 24 Apr 2025
Multi-Level Optimal Transport for Universal Cross-Tokenizer Knowledge Distillation on Language ModelsAAAI Conference on Artificial Intelligence (AAAI), 2024 Xiao Cui Mo Zhu Yulei Qin Liang Xie Wengang Zhou Haoyang Li 411 21 0 19 Dec 2024