English
Related papers

Related papers: Safety Alignment as Continual Learning: Mitigating…

200 papers

Multi-task post-training of large language models (LLMs) is typically performed by mixing datasets from different tasks and optimizing them jointly. This approach implicitly assumes that all tasks contribute gradients of similar magnitudes;…

The increasing complexity of neural networks poses significant challenges for democratizing FL on resource?constrained client devices. Parallel split learning (PSL) has emerged as a promising solution by offloading substantial computing…

Machine Learning · Computer Science 2026-03-20 Zheng Lin , Ons Aouedi , Wei Ni , Symeon Chatzinotas , Xianhao Chen

Large language models (LLMs) have emerged as powerful tools for addressing a wide range of general inquiries and tasks. Despite this, fine-tuning aligned LLMs on smaller, domain-specific datasets, critical to adapting them to specialized…

Artificial Intelligence · Computer Science 2025-02-04 Guanlin Li , Kangjie Chen , Shangwei Guo , Jie Zhang , Han Qiu , Chao Zhang , Guoyin Wang , Tianwei Zhang , Jiwei Li

Adaptive optimization methods (such as Adam) play a major role in LLM pretraining, significantly outperforming Gradient Descent (GD). Recent studies have proposed new smoothness assumptions on the loss function to explain the advantages of…

Machine Learning · Computer Science 2025-12-02 Robin Yadav , Shuo Xie , Tianhao Wang , Zhiyuan Li

This paper focuses on the alignment of flow matching models with human preferences. A promising way is fine-tuning by directly backpropagating reward gradients through the differentiable generation process of flow matching. However,…

Computer Vision and Pattern Recognition · Computer Science 2026-05-05 Zhanhao Liang , Tao Yang , Jie Wu , Chengjian Feng , Liang Zheng

Recent breakthroughs in preference alignment have significantly improved Large Language Models' ability to generate texts that align with human preferences and values. However, current alignment metrics typically emphasize the post-hoc…

Computation and Language · Computer Science 2024-10-15 Mingye Zhu , Yi Liu , Quan Wang , Junbo Guo , Zhendong Mao

While large machine learning models have shown remarkable performance in various domains, their training typically requires iterating for many passes over the training data. However, due to computational and memory constraints and potential…

Machine Learning · Computer Science 2025-11-04 Youngjae Min , Namhoon Cho , Navid Azizan

A fundamental requirement for intelligent systems is the ability to learn continuously under changing environments. However, models trained in this regime often suffer from catastrophic forgetting. Leveraging pre-trained models has recently…

Artificial Intelligence · Computer Science 2026-03-12 Tung Tran , Danilo Vasconcellos Vargas , Khoat Than

Although multi-task learning (MTL) has been a preferred approach and successfully applied in many real-world scenarios, MTL models are not guaranteed to outperform single-task models on all tasks mainly due to the negative effects of…

Machine Learning · Computer Science 2025-03-06 Shijie Zhu , Hui Zhao , Tianshu Wu , Pengjie Wang , Hongbo Deng , Jian Xu , Bo Zheng

The performance of an optimizer on large-scale deep learning models depends critically on fine-tuning the learning rate, often requiring an extensive grid search over base learning rates, schedules, and other hyperparameters. In this paper,…

Machine Learning · Computer Science 2025-06-11 Ruichen Jiang , Ali Kavis , Aryan Mokhtari

Current safety alignment techniques for large language models (LLMs) face two key challenges: (1) under-generalization, which leaves models vulnerable to novel jailbreak attacks, and (2) over-alignment, which leads to the excessive refusal…

Computation and Language · Computer Science 2025-04-15 Yutao Mou , Yuxiao Luo , Shikun Zhang , Wei Ye

Ensuring alignment with human preferences is a crucial characteristic of large language models (LLMs). Presently, the primary alignment methods, RLHF and DPO, require extensive human annotation, which is expensive despite their efficacy.…

Computation and Language · Computer Science 2024-04-16 Haotian Luo

Post-training alignment of large language models (LLMs) relies on large-scale human annotations guided by policy specifications that change over time. Cultural shifts, value reinterpretations, and regulatory or industrial updates make…

Computation and Language · Computer Science 2026-05-12 Aakash Sen Sharma , Debdeep Sanyal , Manodeep Ray , Vivek Srivastava , Shirish Karande , Murari Mandal

Many modern learning tasks involve fitting nonlinear models to data which are trained in an overparameterized regime where the parameters of the model exceed the size of the training dataset. Due to this overparameterization, the training…

Machine Learning · Computer Science 2018-12-27 Samet Oymak , Mahdi Soltanolkotabi

The safety mechanisms of large language models (LLMs) exhibit notable fragility, as even fine-tuning on datasets without harmful content may still undermine their safety capabilities. Meanwhile, existing safety alignment methods…

Computers and Society · Computer Science 2026-02-03 Guanghao Zhou , Panjia Qiu , Cen Chen , Hongyu Li , Mingyuan Chu , Xin Zhang , Jun Zhou

Alignment tuning has enabled large language models to excel in reasoning, instruction-following, and minimizing harmful generations. However, despite their widespread deployment, these models exhibit a monolingual bias, raising concerns…

Computation and Language · Computer Science 2025-04-04 Nikhil Verma , Manasa Bharadwaj

Vision-Language Models require efficient adaptation to continually emerging downstream tasks. While Parameter-Efficient Fine-Tuning mitigates catastrophic forgetting, assigning isolated modules per task leads to parameter explosion.…

Computer Vision and Pattern Recognition · Computer Science 2026-05-20 Xuezhi Cui , Dongbo Zhou , Wang Guo , Zeyuan Wang , Ziyu Li , Gaozhi Zhou , Xian Li , Ling Zhao , Wentao Yang , Chao Tao , Haifeng Li

Open-set domain adaptation (OSDA) has gained considerable attention in many visual recognition tasks. However, most existing OSDA approaches are limited due to three main reasons, including: (1) the lack of essential theoretical analysis of…

Computer Vision and Pattern Recognition · Computer Science 2023-01-25 Yadan Luo , Zijian Wang , Zhuoxiao Chen , Zi Huang , Mahsa Baktashmotlagh

Existing reinforcement learning (RL)-based post-training methods for large language models have advanced rapidly, yet their design has largely been guided by heuristics rather than systematic theoretical principles. This gap limits our…

Machine Learning · Statistics 2026-01-16 Zixun Huang , Jiayi Sheng , Zeyu Zheng

Gradient inversion attacks reveal that private training text can be reconstructed from shared gradients, posing a privacy risk to large language models (LLMs). While prior methods perform well in small-batch settings, scaling to larger…

Machine Learning · Computer Science 2026-03-18 Yibo Li , Qiongxiu Li