English
Related papers

Related papers: Revisiting Robustness for LLM Safety Alignment via…

200 papers

The inherent safety alignment of Large Language Models (LLMs) is prone to erosion during fine-tuning, even when using seemingly innocuous datasets. While existing defenses attempt to mitigate this via data selection, they typically rely on…

Machine Learning · Computer Science 2026-01-13 Haozhong Wang , Zhuo Li , Yibo Yang , He Zhao , Hongyuan Zha , Dandan Guo

Direct Preference Optimization (DPO) has become a popular method for fine-tuning large language models (LLMs) due to its stability and simplicity. However, it is also known to be sensitive to noise in the data and prone to overfitting.…

Machine Learning · Computer Science 2025-10-28 Cheol Woo Kim , Shresth Verma , Mauricio Tec , Milind Tambe

Reinforcement learning (RL) plays an increasingly important role in enhancing the reasoning capabilities of large language models (LLMs), yet stable and performant policy optimization remains challenging. Token-level importance ratios often…

Machine Learning · Computer Science 2025-12-02 Chang Gao , Chujie Zheng , Xiong-Hui Chen , Kai Dang , Shixuan Liu , Bowen Yu , An Yang , Shuai Bai , Jingren Zhou , Junyang Lin

Existing Reinforcement Learning with Verifiable Rewards (RLVR) algorithms, such as GRPO, rely on rigid, uniform, and symmetric trust region mechanisms that are fundamentally misaligned with the complex optimization dynamics of Large…

Machine Learning · Computer Science 2026-04-21 Xiaoliang Fu , Jiaye Lin , Yangyi Fang , Binbin Zheng , Chaowen Hu , Zekai Shao , Cong Qin , Lu Pan , Ke Zeng , Xunliang Cai

Fine-tuning LLMs on benign data can still degrade alignment and adversarial robustness, yet direct analysis of the role of fine-tuning objectives in shaping these safety outcomes remain limited. We present a controlled comparison of six…

Computation and Language · Computer Science 2026-01-21 Daniel Vennemeyer , Punya Syon Pandey , Phan Anh Duong , Michael Umeokoli , Samuel Ratnam

Aligning large language models with human preferences must balance two competing goals: responding helpfully to legitimate requests and reliably refusing harmful ones. Most preference-based safety alignment methods collapse safety into a…

Computation and Language · Computer Science 2026-04-22 Tiankai Yang , Yi Nian , Xinyuan Li , Ruiyao Xu , Kaize Ding , Yue Zhao

Large Vision-Language Models (LVLMs), such as GPT-4o and LLaVA, have recently witnessed remarkable advancements and are increasingly being deployed in real-world applications. However, inheriting the sensitivity of visual neural networks,…

Computer Vision and Pattern Recognition · Computer Science 2025-04-03 Chaohu Liu , Tianyi Gui , Yu Liu , Linli Xu

Despite recent advances in Large Video Language Models (LVLMs), they still struggle with fine-grained temporal understanding, hallucinate, and often make simple mistakes on even simple video question-answering tasks, all of which pose…

Computer Vision and Pattern Recognition · Computer Science 2025-09-30 Pritam Sarkar , Ali Etemad

Large language models (LLMs) require careful alignment to balance competing objectives - factuality, safety, conciseness, proactivity, and diversity. Existing studies focus on individual techniques or specific dimensions, lacking a holistic…

Machine Learning · Computer Science 2025-09-17 Denis Janiak , Julia Moska , Dawid Motyka , Karolina Seweryn , Paweł Walkowiak , Bartosz Żuk , Arkadiusz Janz

As large language models (LLMs) are increasingly applied across various domains, enhancing safety while maintaining the helpfulness of LLMs has become a critical challenge. Recent studies solve this problem through safety-constrained online…

Computation and Language · Computer Science 2025-06-04 Yupeng Qi , Ziyu Lyu , Min Yang , Yanlin Wang , Lu Bai , Lixin Cui

Large Reasoning Models (LRMs) have achieved tremendous success with their chain-of-thought (CoT) reasoning, yet also face safety issues similar to those of basic language models. In particular, while algorithms are designed to guide them to…

Machine Learning · Computer Science 2026-02-05 Zeming Wei , Qiaosheng Zhang , Xia Hu , Xingcheng Xu

Existing multi-objective preference alignment methods for large language models (LLMs) face limitations: (1) the inability to effectively balance various preference dimensions, and (2) reliance on auxiliary reward/reference models…

Machine Learning · Computer Science 2025-06-10 Qi Liu , Jingqing Ruan , Hao Li , Haodong Zhao , Desheng Wang , Jiansong Chen , Wan Guanglu , Xunliang Cai , Zhi Zheng , Tong Xu

This study addresses the challenge of noise in training datasets for Direct Preference Optimization (DPO), a method for aligning Large Language Models (LLMs) with human preferences. We categorize noise into pointwise noise, which includes…

Machine Learning · Computer Science 2025-04-21 Junkang Wu , Yuexiang Xie , Zhengyi Yang , Jiancan Wu , Jiawei Chen , Jinyang Gao , Bolin Ding , Xiang Wang , Xiangnan He

Improving and understanding the training dynamics and reasoning of Large Language Models (LLMs) has become essential for their deployment in AI-based security tools, such as software vulnerability detection. In this work, we present an…

Cryptography and Security · Computer Science 2025-07-08 Marco Simoni , Aleksandar Fontana , Giulio Rossolini , Andrea Saracino

The rapidly increasing capabilities of large language models (LLMs) raise an urgent need to align AI systems with diverse human preferences to simultaneously enhance their usefulness and safety, despite the often conflicting nature of these…

Machine Learning · Computer Science 2024-03-06 Zixuan Liu , Xiaolin Sun , Zizhan Zheng

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

Neural metrics for machine translation (MT) evaluation have become increasingly prominent due to their superior correlation with human judgments compared to traditional lexical metrics. Researchers have therefore utilized neural metrics…

Computation and Language · Computer Science 2025-11-21 Hippolyte Gisserot-Boukhlef , Ricardo Rei , Emmanuel Malherbe , Céline Hudelot , Pierre Colombo , Nuno M. Guerreiro

Group Relative Policy Optimization(GRPO) has become a cornerstone of modern reinforcement learning alignment, prized for its efficacy in foregoing an explicit value-critic by leveraging reward normalization across sampled trajectory…

Computation and Language · Computer Science 2026-05-29 Redacted by arXiv

Training large language models (LLMs) with reinforcement learning (RL) methods such as PPO and GRPO commonly relies on ratio clipping to stabilise updates. While effective at preventing instability, clipping discards information, introduces…

Machine Learning · Computer Science 2026-02-02 Madeleine Dwyer , Adam Sobey , Adriane Chapman

Multimodal large language models (MLLMs) have demonstrated impressive reasoning and instruction-following capabilities, yet their expanded modality space introduces new compositional safety risks that emerge from complex text-image…

Cryptography and Security · Computer Science 2025-11-18 Xuankun Rong , Wenke Huang , Tingfeng Wang , Daiguo Zhou , Bo Du , Mang Ye