English
Related papers

Related papers: Model Unlearning via Sparse Autoencoder Subspace G…

200 papers

Large language models (LLMs) acquire a large amount of knowledge through pre-training on vast and diverse corpora. While this endows LLMs with strong capabilities in generation and reasoning, it amplifies risks associated with sensitive,…

Cryptography and Security · Computer Science 2026-02-25 Ce Fang , Zhikun Zhang , Min Chen , Qing Liu , Lu Zhou , Zhe Liu , Yunjun Gao

Machine unlearning for large language models (LLMs) aims to remove undesired data, knowledge, and behaviors (e.g., for safety, privacy, or copyright) while preserving useful model capabilities. Despite rapid progress over the past two…

Machine Learning · Computer Science 2025-10-10 Chongyu Fan , Changsheng Wang , Yancheng Huang , Soumyadeep Pal , Sijia Liu

Large language models inevitably retain sensitive information, defined as inputs that may induce harmful generations, due to training on massive web corpora, raising concerns for privacy and safety. Existing machine unlearning methods…

Machine Learning · Computer Science 2026-05-21 Yujie Lin , Chengyi Yang , Zhishang Xiang , Yiping Song , Jinsong Su

Sparse autoencoders (SAEs) are increasingly used to extract activation directions for inference-time steering, but their standard sparsity objective treats latent features as independent. This prior can be poorly matched to high-level…

Machine Learning · Computer Science 2026-05-18 Jehyeok Yeon , Federico Cinus , Yifan Wu , Luca Luceri

Although large-scale self-supervised learning (SSL) models like WavLM have achieved state-of-the-art performance in speech processing, their significant size impedes deployment on resource-constrained devices. While structured pruning is a…

Audio and Speech Processing · Electrical Eng. & Systems 2025-11-11 Junyi Peng , Lin Zhang , Jiangyu Han , Oldřich Plchot , Johan Rohdin , Themos Stafylakis , Shuai Wang , Jan Černocký

The security of biomedical Multimodal Large Language Models (MLLMs) has attracted increasing attention. However, training samples easily contain private information and incorrect knowledge that are difficult to detect, potentially leading…

Computer Vision and Pattern Recognition · Computer Science 2025-08-07 Dunyuan Xu , Xikai Yang , Yaoqian Li , Jinpeng Li , Pheng-Ann Heng

Large language models (LLMs) are now ubiquitous in user-facing applications, yet they still generate undesirable toxic outputs, including profanity, vulgarity, and derogatory remarks. Although numerous detoxification methods exist, most…

Computation and Language · Computer Science 2025-10-24 Agam Goyal , Vedant Rathi , William Yeh , Yian Wang , Yuen Chen , Hari Sundaram

Self-supervised learning (SSL) has emerged as a powerful strategy for representation learning under limited annotation regimes, yet its effectiveness remains highly sensitive to many factors, especially the nature of the target task. In…

Computer Vision and Pattern Recognition · Computer Science 2026-01-27 Jorge Quesada , Ghassan AlRegib

The advent of Large Language Models (LLMs) has revolutionized natural language processing, enabling advanced understanding and reasoning capabilities across a variety of tasks. Fine-tuning these models for specific domains, particularly…

Artificial Intelligence · Computer Science 2025-04-21 Chenlu Ding , Jiancan Wu , Yancheng Yuan , Jinda Lu , Kai Zhang , Alex Su , Xiang Wang , Xiangnan He

Given the prevalence of large language models (LLMs) and the prohibitive cost of training these models from scratch, dynamically forgetting specific knowledge e.g., private or proprietary, without retraining the model has become an…

Computation and Language · Computer Science 2024-08-09 Tyler Lizzo , Larry Heck

With the integration of image modality, the semantic space of multimodal large language models (MLLMs) is more complex than text-only models, making their interpretability more challenging and their alignment less stable, particularly…

Machine Learning · Computer Science 2025-06-18 Hantao Lou , Changye Li , Jiaming Ji , Yaodong Yang

Once users have shared their data online, it is generally difficult for them to revoke access and ask for the data to be deleted. Machine learning (ML) exacerbates this problem because any model trained with said data may have memorized it,…

Large-scale pre-trained vision-language models like CLIP demonstrate remarkable zero-shot performance across diverse tasks. However, fine-tuning these models to improve downstream performance often degrades robustness against distribution…

Computer Vision and Pattern Recognition · Computer Science 2026-05-18 Fabian Morelli , Arnas Uselis , Ankit Sonthalia , Seong Joon Oh

Neural models have yielded state-of-the-art results in deciphering spoken language understanding (SLU) problems; however, these models require a significant amount of domain-specific labeled examples for training, which is prohibitively…

Computation and Language · Computer Science 2020-10-12 Jin Cao , Jun Wang , Wael Hamza , Kelly Vanee , Shang-Wen Li

Language models (LMs) are trained on vast amounts of text data, which may include private and copyrighted content. Data owners may request the removal of their data from a trained model due to privacy or copyright concerns. However, exactly…

Computation and Language · Computer Science 2024-07-16 Weijia Shi , Jaechan Lee , Yangsibo Huang , Sadhika Malladi , Jieyu Zhao , Ari Holtzman , Daogao Liu , Luke Zettlemoyer , Noah A. Smith , Chiyuan Zhang

Sparse Autoencoders (SAEs) are powerful tools for interpreting neural representations, yet their use in audio remains underexplored. We train SAEs across all encoder layers of Whisper and HuBERT, provide an extensive evaluation of their…

As large language models (LLMs) are increasingly deployed in real-world systems, they must support post-hoc removal of specific content to meet privacy and governance requirements. This motivates selective unlearning, which suppresses…

Computation and Language · Computer Science 2026-05-28 Chenchen Tan , Xinghao Li , Shujie Cui , Youyang Qu , Cunjian Chen , Longxiang Gao

Machine unlearning has emerged as a critical capability for addressing privacy, safety, and regulatory concerns in large language models (LLMs). Existing methods operate at the sequence level, applying uniform updates across all tokens…

Computation and Language · Computer Science 2026-05-07 Jiawei Wu , Doudou Zhou

Sparse autoencoders (SAEs) are widely used in mechanistic interpretability to project LLM activations onto sparse latent spaces. However, sparsity alone is an imperfect proxy for interpretability, and current training objectives often…

Machine Learning · Computer Science 2026-04-09 Vivek Narayanaswamy , Kowshik Thopalli , Bhavya Kailkhura , Wesam Sakla

In this paper, we introduce \textbf{agentic unlearning} which removes specified information from both model parameters and persistent memory in agents with closed-loop interaction. Existing unlearning methods target parameters alone,…

Machine Learning · Computer Science 2026-03-03 Bin Wang , Fan Wang , Pingping Wang , Jinyu Cong , Yang Yu , Yilong Yin , Zhongyi Han , Benzheng Wei