English
Related papers

Related papers: Backtracking Improves Generation Safety

200 papers

Backdoor attack introduces artificial vulnerabilities into the model by poisoning a subset of the training data via injecting triggers and modifying labels. Various trigger design strategies have been explored to attack text classifiers,…

Computation and Language · Computer Science 2021-09-23 Zichao Li , Dheeraj Mekala , Chengyu Dong , Jingbo Shang

Large Language Model (LLM) alignment remains vulnerable to jailbreak attacks that elicit unsafe responses, motivating pre-model and post-model guards. Pre-model guards audit the safety of prompts before invoking target models. However,…

Cryptography and Security · Computer Science 2026-05-20 Hongyu Cai , Arjun Arunasalam , Yiming Liang , Antonio Bianchi , Z. Berkay Celik

Conversational models that are generative and open-domain are particularly susceptible to generating unsafe content since they are trained on web-based social data. Prior approaches to mitigating this issue have drawbacks, such as…

Computation and Language · Computer Science 2023-04-26 Leila Khalatbari , Yejin Bang , Dan Su , Willy Chung , Saeed Ghadimi , Hossein Sameti , Pascale Fung

The jailbreak attack can bypass the safety measures of a Large Language Model (LLM), generating harmful content. This misuse of LLM has led to negative societal consequences. Currently, there are two main approaches to address jailbreak…

Computation and Language · Computer Science 2024-03-25 Zezhong Wang , Fangkai Yang , Lu Wang , Pu Zhao , Hongru Wang , Liang Chen , Qingwei Lin , Kam-Fai Wong

Current methods for content safety in Large Language Models (LLMs), such as Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF), often rely on multi-stage training pipelines and lack fine-grained,…

Computation and Language · Computer Science 2026-01-21 Jianfeng Si , Lin Sun , Zhewen Tan , Xiangzheng Zhang

Alignment training has tradeoffs: it helps language models (LMs) gain in reasoning and instruction following but might lose out on skills such as creativity and calibration, where unaligned base models are better at. We aim to make the best…

Computation and Language · Computer Science 2025-10-14 Shangbin Feng , Wenhao Yu , Yike Wang , Hongming Zhang , Yulia Tsvetkov , Dong Yu

As the influence of large language models (LLMs) spans across global communities, their safety challenges in multilingual settings become paramount for alignment research. This paper examines the variations in safety challenges faced by…

Computation and Language · Computer Science 2024-01-25 Lingfeng Shen , Weiting Tan , Sihao Chen , Yunmo Chen , Jingyu Zhang , Haoran Xu , Boyuan Zheng , Philipp Koehn , Daniel Khashabi

Backdoor attacks pose a serious threat to the security of large language models (LLMs), causing them to exhibit anomalous behavior under specific trigger conditions. The design of backdoor triggers has evolved from fixed triggers to dynamic…

Cryptography and Security · Computer Science 2026-04-15 Haotian Jin , Yang Li , Haihui Fan , Lin Shen , Xiangfang Li , Bo Li

The frustratingly fragile nature of neural network models make current natural language generation (NLG) systems prone to backdoor attacks and generate malicious sequences that could be sexist or offensive. Unfortunately, little effort has…

Computation and Language · Computer Science 2023-10-10 Xiaofei Sun , Xiaoya Li , Yuxian Meng , Xiang Ao , Lingjuan Lyu , Jiwei Li , Tianwei Zhang

Estimating uncertainty in Large Language Models (LLMs) is important for properly evaluating LLMs, and ensuring safety for users. However, prior approaches to uncertainty estimation focus on the final answer in generated text, ignoring…

Computation and Language · Computer Science 2024-12-12 Eric Bigelow , Ari Holtzman , Hidenori Tanaka , Tomer Ullman

Large language models (LLMs) employ safety mechanisms to prevent harmful outputs, yet these defenses primarily rely on semantic pattern matching. We show that encoding harmful prompts as coherent mathematical problems -- using formalisms…

Cryptography and Security · Computer Science 2026-05-06 Haoyu Zhang , Mohammad Zandsalimy , Shanu Sushmita

Large language models (LLMs) have demonstrated immense utility across various industries. However, as LLMs advance, the risk of harmful outputs increases due to incorrect or malicious instruction prompts. While current methods effectively…

Computation and Language · Computer Science 2025-06-19 Xinyi Zeng , Yuying Shang , Jiawei Chen , Jingyuan Zhang , Yu Tian

With the ability to generate high-quality images, text-to-image (T2I) models can be exploited for creating inappropriate content. To prevent misuse, existing safety measures are either based on text blacklists, which can be easily…

Computer Vision and Pattern Recognition · Computer Science 2024-08-20 Runtao Liu , Ashkan Khakzar , Jindong Gu , Qifeng Chen , Philip Torr , Fabio Pizzati

We study the text generation task under the approach of pre-trained language models (PLMs). Typically, an auto-regressive (AR) method is adopted for generating texts in a token-by-token manner. Despite many advantages of AR generation, it…

Computation and Language · Computer Science 2022-10-31 Junyi Li , Tianyi Tang , Wayne Xin Zhao , Jian-Yun Nie , Ji-Rong Wen

This paper argues that generating output tokens is more effective than using pooled representations for prediction tasks because token-level generation retains more mutual information. Since LLMs are trained on massive text corpora using…

Jailbreak attacks pose a serious threat to the safety of Large Language Models (LLMs) by crafting adversarial prompts that bypass alignment mechanisms, causing the models to produce harmful, restricted, or biased content. In this paper, we…

Machine Learning · Computer Science 2025-08-22 Xiangman Li , Xiaodong Wu , Qi Li , Jianbing Ni , Rongxing Lu

Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning and text generation. However, these models can inadvertently generate unsafe or biased responses when prompted with problematic inputs, raising…

Computation and Language · Computer Science 2024-12-03 Avinash Amballa , Durga Sandeep Saluru , Gayathri Akkinapalli , Abhishek Sureddy , Akshay Kumar Sureddy

Training multimodal generative models on large, uncurated datasets can result in users being exposed to harmful, unsafe and controversial or culturally-inappropriate outputs. While model editing has been proposed to remove or filter…

Computer Vision and Pattern Recognition · Computer Science 2025-03-06 Jordan Vice , Naveed Akhtar , Mubarak Shah , Richard Hartley , Ajmal Mian

Language models are commonly fine-tuned for safety alignment to refuse harmful prompts. One approach fine-tunes them to generate categorical refusal tokens that distinguish different refusal types before responding. In this work, we…

Artificial Intelligence · Computer Science 2026-03-17 Rishab Alagharu , Ishneet Sukhvinder Singh , Shaibi Shamsudeen , Zhen Wu , Ashwinee Panda

Diffusion Language Models (DLMs) decode multiple tokens in parallel, but aggressive multi-token decoding amplifies cross-token dependency errors and can sharply degrade generation quality. We propose BackPlay, a frozen-backbone…

Machine Learning · Computer Science 2026-04-24 Liming Liu , Binxuan Huang , Zixuan Zhang , Xin Liu , Bing Yin , Tuo Zhao
‹ Prev 1 4 5 6 7 8 10 Next ›