English
Related papers

Related papers: Prompt Attacks Reveal Superficial Knowledge Remova…

200 papers

Reasoning-enhanced large language models (RLLMs), whether explicitly trained for reasoning or prompted via chain-of-thought (CoT), have achieved state-of-the-art performance on many complex reasoning tasks. However, we uncover a surprising…

Computation and Language · Computer Science 2025-09-03 Xiaomin Li , Zhou Yu , Zhiwei Zhang , Xupeng Chen , Ziji Zhang , Yingying Zhuang , Narayanan Sadagopan , Anurag Beniwal

How can we extend a pre-trained model to many language understanding tasks, without labeled or additional unlabeled data? Pre-trained language models (PLMs) have been effective for a wide range of NLP tasks. However, existing approaches…

Computation and Language · Computer Science 2023-05-29 Xuandong Zhao , Siqi Ouyang , Zhiguo Yu , Ming Wu , Lei Li

Machine unlearning algorithms, designed for selective removal of training data from models, have emerged as a promising approach to growing privacy concerns. In this work, we expose a critical yet underexplored vulnerability in the…

Cryptography and Security · Computer Science 2024-10-15 Yangsibo Huang , Daogao Liu , Lynn Chua , Badih Ghazi , Pritish Kamath , Ravi Kumar , Pasin Manurangsi , Milad Nasr , Amer Sinha , Chiyuan Zhang

We show that existing unsupervised methods on large language model (LLM) activations do not discover knowledge -- instead they seem to discover whatever feature of the activations is most prominent. The idea behind unsupervised knowledge…

Machine Learning · Computer Science 2023-12-19 Sebastian Farquhar , Vikrant Varma , Zachary Kenton , Johannes Gasteiger , Vladimir Mikulik , Rohin Shah

Large Language Models (LLMs) have demonstrated remarkable performance across a wide range of applications, e.g., medical question-answering, mathematical sciences, and code generation. However, they also exhibit inherent limitations, such…

Cryptography and Security · Computer Science 2025-06-23 Yang Jiao , Xiaodong Wang , Kai Yang

Large language models (LLMs) store vast amounts of information, making them powerful yet raising privacy and safety concerns when selective knowledge removal is required. Existing unlearning strategies, ranging from gradient-based…

Computation and Language · Computer Science 2025-06-02 Xu Wang , Zihao Li , Benyou Wang , Yan Hu , Difan Zou

Here, we show that current LLM unlearning methods inherently reduce models' robustness, causing them to misbehave even when a single non-adversarial forget-token is present in the retain-query. Toward understanding underlying causes, we…

Computation and Language · Computer Science 2026-04-21 Dang Huu-Tien , Hoang Thanh-Tung , Anh Bui , Minh-Phuong Nguyen , Le-Minh Nguyen , Naoya Inoue

Pretrained language models memorize vast amounts of information, including private and copyrighted data, raising significant safety concerns. Retraining these models after excluding sensitive data is prohibitively expensive, making machine…

Computation and Language · Computer Science 2024-10-04 Minseok Choi , Kyunghyun Min , Jaegul Choo

Machine unlearning enables pre-trained models to eliminate the effects of partial training samples. Previous research has mainly focused on proposing efficient unlearning strategies. However, the verification of machine unlearning, or in…

Cryptography and Security · Computer Science 2024-06-18 Heng Xu , Tianqing Zhu , Lefeng Zhang , Wanlei Zhou

Large language model (LLM) unlearning has emerged as a crucial post-hoc mechanism for privacy protection and AI safety, yet auditing whether target knowledge is truly erased remains challenging. Existing output-level metrics fail to detect…

Computation and Language · Computer Science 2026-05-26 Jaeung Lee , Dohyun Kim , Jaemin Jo

Jailbreaking attacks can enable Large Language Models (LLMs) to bypass the safeguard and generate harmful content. Existing jailbreaking defense methods have failed to address the fundamental issue that harmful knowledge resides within the…

Computation and Language · Computer Science 2024-07-04 Weikai Lu , Ziqian Zeng , Jianwei Wang , Zhengdong Lu , Zelin Chen , Huiping Zhuang , Cen Chen

Recent studies demonstrate that Large Language Models (LLMs) are vulnerable to different prompt-based attacks, generating harmful content or sensitive information. Both closed-source and open-source LLMs are underinvestigated for these…

Cryptography and Security · Computer Science 2025-05-21 Jiawen Wang , Pritha Gupta , Ivan Habernal , Eyke Hüllermeier

Machine unlearning is the process of efficiently removing specific information from a trained machine learning model without retraining from scratch. Existing unlearning methods, which often provide provable guarantees, typically involve…

Machine Learning · Computer Science 2026-02-04 Somnath Basu Roy Chowdhury , Rahul Kidambi , Avinava Dubey , David Wang , Gokhan Mergen , Amr Ahmed , Aranyak Mehta

This work presents LURK (Latent UnleaRned Knowledge), a novel framework that probes for hidden retained knowledge in unlearned LLMs through adversarial suffix prompting. LURK automatically generates adversarial prompt suffixes designed to…

Computation and Language · Computer Science 2025-05-26 Bang Trinh Tran To , Thai Le

Large language models are increasingly used for vulnerability detection, yet their reliability under different prompt formulations remains uncharacterized. We present PromptAudit, a controlled evaluation framework that isolates prompt…

Machine Learning · Computer Science 2026-05-26 Steffen J. Camarato , Yahya Hmaiti , Mandana Ghadamian , David Mohaisen

Multimodal contrastive learning uses various data modalities to create high-quality features, but its reliance on extensive data sources on the Internet makes it vulnerable to backdoor attacks. These attacks insert malicious behaviors…

Cryptography and Security · Computer Science 2024-10-01 Kuanrong Liu , Siyuan Liang , Jiawei Liang , Pengwen Dai , Xiaochun Cao

This study reveals a previously unexplored vulnerability in the safety alignment of Large Language Models (LLMs). Existing aligned LLMs predominantly respond to unsafe queries with refusals, which often begin with a fixed set of prefixes…

Cryptography and Security · Computer Science 2026-01-28 Yangyang Guo , Ziwei Xu , Si Liu , Zhiming Zheng , Mohan Kankanhalli

Machine learning models are vulnerable to adversarial attacks, including attacks that leak information about the model's training data. There has recently been an increase in interest about how to best address privacy concerns, especially…

Machine Learning · Computer Science 2024-05-30 Keltin Grimes , Collin Abidi , Cole Frank , Shannon Gallagher

In many practical applications, large language models (LLMs) need to acquire new knowledge not present in their pre-training data. Efficiently leveraging this knowledge usually relies on supervised fine-tuning or retrieval-augmented…

Computation and Language · Computer Science 2025-08-08 Kalle Kujanpää , Pekka Marttinen , Harri Valpola , Alexander Ilin

Machine unlearning has become an important area of research due to an increasing need for machine learning (ML) applications to comply with the emerging data privacy regulations. It facilitates the provision for removal of certain set or…

Machine Learning · Computer Science 2023-06-01 Vikram S Chundawat , Ayush K Tarun , Murari Mandal , Mohan Kankanhalli