English
Related papers

Related papers: Refusal Before Decoding: Detecting and Exploiting …

200 papers

Pre-trained Transformers, through in-context learning (ICL), have demonstrated exceptional capabilities to adapt to new tasks using example prompts \textit{without model update}. Transformer-based wireless receivers, where prompts consist…

Information Theory · Computer Science 2024-11-13 Li Fan , Jing Yang , Cong Shen

LLM prompting is widely used for naturally stated tasks, yet it is unreliable it may succeed on a few test cases but fail at deployment time. We study performance prediction: given a program, either symbolic (e.g. Python) or a prompt…

Machine Learning · Computer Science 2026-05-22 Chengqi Zheng , Keya Hu , Shuzhi Liu , Tao Wu , Kevin Ellis , Yewen Pu

Backdoor unalignment attacks against Large Language Models (LLMs) enable the stealthy compromise of safety alignment using a hidden trigger while evading normal safety auditing. These attacks pose significant threats to the applications of…

Cryptography and Security · Computer Science 2025-06-23 Biao Yi , Tiansheng Huang , Sishuo Chen , Tong Li , Zheli Liu , Zhixuan Chu , Yiming Li

We propose a method to teach multiple large language models (LLM) to collaborate by interleaving their generations at the token level. We model the decision of which LLM generates the next token as a latent variable. By optimizing the…

Computation and Language · Computer Science 2024-08-28 Shannon Zejiang Shen , Hunter Lang , Bailin Wang , Yoon Kim , David Sontag

Large language models (LLMs) showcase impressive reasoning capabilities when coupled with Chain-of-Thought (CoT) prompting. However, the robustness of this approach warrants further investigation. In this paper, we introduce a novel…

Computation and Language · Computer Science 2024-06-03 Rongwu Xu , Zehan Qi , Wei Xu

Agents powered by large language models have shown remarkable abilities in solving complex tasks. However, most agent systems remain reactive, limiting their effectiveness in scenarios requiring foresight and autonomous decision-making. In…

The capabilities of Large Language Models (LLMs) are routinely evaluated by other LLMs trained to predict human preferences. This framework--known as LLM-as-a-judge--is highly scalable and relatively low cost. However, it is also vulnerable…

Computation and Language · Computer Science 2026-02-03 Lisa Alazraki , Tan Yi-Chern , Jon Ander Campos , Maximilian Mozes , Marek Rei , Max Bartolo

Models need appropriate inductive biases to effectively learn from small amounts of data and generalize systematically outside of the training distribution. While Transformers are highly versatile and powerful, they can still benefit from…

Computation and Language · Computer Science 2024-07-08 Matthias Lindemann , Alexander Koller , Ivan Titov

Existing large language models (LLMs) occasionally generate plausible yet factually incorrect responses, known as hallucinations. Two main approaches have been proposed to mitigate hallucinations: retrieval-augmented language models (RALMs)…

Computation and Language · Computer Science 2025-11-19 Youchao Zhou , Heyan Huang , Yicheng Liu , Rui Dai , Xinglin Wang , Xingchen Zhang , Shumin Shi , Yang Deng

Language models (LMs) are increasingly used to build agents that can act autonomously to achieve goals. During this automatic process, agents need to take a series of actions, some of which might lead to severe consequences if incorrect…

Computation and Language · Computer Science 2025-10-01 Cheng-Kuang Wu , Zhi Rui Tam , Chieh-Yen Lin , Yun-Nung Chen , Hung-yi Lee

Can linearly decodable failure signals in LLM hidden states be leveraged to correct those failures? We investigate this classification-correction gap via Overthinking (OT)--a stable behavioral regime (Jaccard >= 0.81, 94% inter-annotator…

Artificial Intelligence · Computer Science 2026-05-08 Ming Liu

Large Language Models (LLMs), despite advances in instruction tuning, often fail to follow complex user instructions. Activation steering techniques aim to mitigate this by manipulating model internals, but have a potential risk of…

Machine Learning · Computer Science 2026-03-10 Minjae Kang , Jaehyung Kim

Large Language Models (LLMs) are becoming a prominent generative AI tool, where the user enters a query and the LLM generates an answer. To reduce harm and misuse, efforts have been made to align these LLMs to human values using advanced…

Cryptography and Security · Computer Science 2024-11-08 Xiaomeng Hu , Pin-Yu Chen , Tsung-Yi Ho

Safety alignment of large language models currently faces a central challenge: existing alignment techniques often prioritize mitigating responses to harmful prompts at the expense of overcautious behavior, leading models to incorrectly…

Although many large language models (LLMs) have been trained to refuse harmful requests, they are still vulnerable to jailbreaking attacks which rewrite the original prompt to conceal its harmful intent. In this paper, we propose a new…

Computation and Language · Computer Science 2024-06-10 Yihan Wang , Zhouxing Shi , Andrew Bai , Cho-Jui Hsieh

The unsupervised Pretraining method has been widely used in aiding human action recognition. However, existing methods focus on reconstructing the already present frames rather than generating frames which happen in future.In this paper, We…

Computer Vision and Pattern Recognition · Computer Science 2017-12-13 Yu Runsheng , Shi Zhenyu , Ma Qiongxiong , Qing Laiyun

Fine-tuning pre-trained Vision-Language Models (VLMs) has shown remarkable capabilities in medical image and textual depiction synergy. Nevertheless, many pre-training datasets are restricted by patient privacy concerns, potentially…

Computer Vision and Pattern Recognition · Computer Science 2025-01-09 Xu Han , Linghao Jin , Xuezhe Ma , Xiaofeng Liu

Jailbreak prompts can trigger harmful completions on aligned LLMs, In accordance, safety steering has been proposed: test-time activation interventions that steer jailbreak activations to trigger refusal while preserving benign utility.…

Cryptography and Security · Computer Science 2026-05-26 Luoyu Chen , Weiqi Wang , Zhiyi Tian , Chenhan Zhang , Feng Wu , Jianhuan Huang , Ahmed Asiri , Shui Yu

Multimodal large language models (MLLMs) excel at multimodal perception and understanding, yet their tendency to generate hallucinated or inaccurate responses undermines their trustworthiness. Existing methods have largely overlooked the…

Computation and Language · Computer Science 2024-12-17 Yuhao Wang , Zhiyuan Zhu , Heyang Liu , Yusheng Liao , Hongcheng Liu , Yanfeng Wang , Yu Wang

System Instructions in Large Language Models (LLMs) are commonly used to enforce safety policies, define agent behavior, and protect sensitive operational context in agentic AI applications. These instructions may contain sensitive…

Cryptography and Security · Computer Science 2026-04-02 Anubhab Sahu , Diptisha Samanta , Reza Soosahabi