CopyShield is a benchmark that compares three copyright‑defense methods—contrastive decoding, Direct Preference Optimization (DPO), and activation intervention—across two large language models (LLaMA‑3.1‑8B and Mistral‑7B‑v0.3). The study uses controlled memorization of five public‑domain books to measure literal leakage, calibrated non‑literal leakage, utility, and degeneracy, finding that each intervention level yields distinct compliance‑utility trade‑offs. Results show contrastive decoding limits degeneracy but hits a suppression floor, DPO nearly eliminates literal leakage yet causes paraphrase‑loop degeneracy, and activation intervention blocks most non‑literal queries before generation, with human evaluation highlighting coherence and perceived copyright risk differences.
By Maryam Alshehyari, Dushyant Singh Chauhan, Samuele Poppi, Martin Takac, Salem Lahlou, Nils Lukas
TextCloak is a reinforcement‑learning driven framework that generates unlearnable examples (UEs) to protect textual data from unauthorized fine‑tuning of large language models (LLMs). It transforms clean text into perturbations that preserve meaning and naturalness while degrading the utility of models trained on them. Experiments on six datasets and nine LLMs show that TextCloak consistently impairs unauthorized fine‑tuning and remains robust across different architectures, training settings, and adaptive attacks.
By Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu
arXiv:2610.31995v1 Announce Type: cross
Abstract: As large language models (LLMs) face growing demands for data removal driven by regulatory compliance, copyright concerns, and privacy protection, ap...
By Mengying Zhang, Derui Wang, Ibrahim Khalil, Bowen Liu, Xiaoyu Xia, Minhui Xue
arXiv:2606. 06286v1 Announce Type: cross Abstract: Large language models can reproduce training data, but existing memorization evaluations mostly measure whether models can be forced to do so, rather than whether they do so under ordinary use.
By Gianluca Barmina, Peter Schneider-Kamp, Lukas Galke Poech
Deep Contrastive Unlearning for Language Models (DeepCUT) is a framework that removes information from fine‑tuned language models by directly optimizing their latent space. It addresses the challenge of machine unlearning in black‑box models, which has been largely overlooked by previous work that only mitigated output effects. Experiments on real‑world datasets show that DeepCUT consistently outperforms baseline methods in both effectiveness and efficiency.
By Estrid He, Tabinda Sarwar, Ibrahim Khalil, Xun Yi, Ke Wang
The paper introduces CopyGen, a lightweight copy-constrained decoding plugin for frozen large language models that improves exact span reproduction. CopyGen decomposes copying into three decisions—when to copy, how far to copy, and when to stop—using adaptive logit adjustments during verification to emit the longest valid prefix. Across three LLM backbones and four benchmark tasks, it boosts average exact match by 35% relative without finetuning and further improves by 4% with SFT, while also accelerating generation.
By Jinghui Zhang, Lang Gao, Zongfang Liu, Ruihong Zeng, Zirui Song, Rui Yan, Kentaro Inui, Xiuying Chen
The paper demonstrates that fine‑tuning large language models on a single author’s works can trigger the models to reproduce large verbatim excerpts from copyrighted books, even when prompted only with semantic descriptions. Experiments on GPT‑4o, Gemini‑2.5‑Pro, and DeepSeek‑V3.1 show up to 85‑90% recall of held‑out books, with spans exceeding 460 words, and this effect generalizes across authors and model providers. The findings suggest that fine‑tuning reactivates latent memorization from pre‑training, revealing a widespread vulnerability in industry models.
By Xinyue Liu, Niloofar Mireshghallah, Jane C. Ginsburg, Tuhin Chakrabarty
arXiv:2602. 18733v2 Announce Type: replace Abstract: Training data leakage from Large Language Models (LLMs) raises serious concerns related to privacy, security, and copyright compliance.
By Trishita Tiwari, Ari Trachtenberg, G. Edward Suh
arXiv:2607. 15232v1 Announce Type: cross Abstract: A tokenizer fixed at the start of pre-training allocates vocabulary in proportion to the pre-training corpus, reflecting the deployment priorities at that time.
By Jimmy T. H. Smith, Tarek Dakhran, Alberto Cabrera, Simon S. Lee, Paul Pak, Aditya Tadimeti, Tim Seyde, Maxime Labonne, Alexander Amini, Mathias Lechner
arXiv:2606. 16244v1 Announce Type: cross Abstract: Large language models routinely generate code with exploitable security flaws.
By Xiaoyun Xu, Lichao Wu, Jona te Lintelo, Siyu Zhang, Stjepan Picek
arXiv:2607. 28862v1 Announce Type: cross Abstract: The rapid development of Large Language Models (LLMs) has led to significant advances across a wide range of language tasks, while simultaneously raising growing concerns about unauthorized data exploitation and privacy leakage.
By Chengshuai Zhao, Pingchuan Ma, Dawei Li, Bohan Jiang, Zhiyuan Yu, Zhen Tan, Huan Liu
The paper introduces PropMe, a framework that evaluates both the capability of large language models (LLMs) to reproduce training data via prefix attacks and their natural propensity to do so without adversarial prompting. Using SimpleTrace to compare every generation against the full indexed training corpus, the authors test models such as Comma, DFM Decoder, and OLMo3 32B across datasets like Common Pile, Dynaword, and Dolma3. Results show that natural sentence prompts produce less training text than prefix attacks, but unprompted or single-word prompts can reproduce as much or more, often involving duplicated licenses, code, and templates; continual pre‑training only modestly reduces verbatim reproduction.
By Gianluca Barmina, Peter Schneider-Kamp, Lukas Galke Poech