中文
相关论文

相关论文: EVOREFUSE: Evolutionary Prompt Optimization for Ev…

200 篇论文

Large language models (LLMs) have shown remarkable performance on various tasks, but existing evaluation benchmarks are often static and insufficient to fully assess their robustness and generalization in realistic scenarios. Prior work…

计算与语言 · 计算机科学 2025-07-01 JiaRu Wu , Mingwei Liu

To circumvent the alignment of large language models (LLMs), current optimization-based adversarial attacks usually craft adversarial prompts by maximizing the likelihood of a so-called affirmative response. An affirmative response is a…

As large language models (LLMs) are deployed in safety-critical settings, it is essential to ensure that their responses comply with safety standards. Prior research has revealed that LLMs often fail to grasp the notion of safe behaviors,…

人工智能 · 计算机科学 2026-03-09 Kartik Sharma , Yiqiao Jin , Vineeth Rakesh , Yingtong Dou , Menghai Pan , Mahashweta Das , Srijan Kumar

Recent advances for few-shot text classification aim to wrap textual inputs with task-specific prompts to cloze questions. By processing them with a masked language model to predict the masked tokens and using a verbalizer that constructs…

计算与语言 · 计算机科学 2023-06-21 Tongtao Ling , Lei Chen , Yutao Lai , Hai-Lin Liu

Large Language Models (LLMs) have achieved remarkable success in natural language processing tasks, but their massive size and computational demands hinder their deployment in resource-constrained environments. Existing model pruning…

计算与语言 · 计算机科学 2025-08-14 Shangyu Wu , Hongchao Du , Ying Xiong , Shuai Chen , Tei-Wei Kuo , Nan Guan , Chun Jason Xue

Automatic prompt engineering aims to enhance the generation quality of large language models (LLMs). Recent works utilize feedbacks generated from erroneous cases to guide the prompt optimization. During inference, they may further retrieve…

计算与语言 · 计算机科学 2025-05-28 Cilin Yan , Jingyun Wang , Lin Zhang , Ruihui Zhao , Xiaopu Wu , Kai Xiong , Qingsong Liu , Guoliang Kang , Yangyang Kang

Considerable research efforts have been devoted to ensuring that large language models (LLMs) align with human values and generate safe text. However, an excessive focus on sensitivity to certain topics can compromise the model's robustness…

计算与语言 · 计算机科学 2023-08-29 Huachuan Qiu , Shuai Zhang , Anqi Li , Hongliang He , Zhenzhong Lan

Evaluating the capability of Large Language Models (LLMs) in following instructions has heavily relied on a powerful LLM as the judge, introducing unresolved biases that deviate the judgments from human judges. In this work, we reevaluate…

计算与语言 · 计算机科学 2025-03-26 Xinxi Lyu , Yizhong Wang , Hannaneh Hajishirzi , Pradeep Dasigi

Large language models (LLMs) frequently produce false refusals, declining benign requests that contain terms resembling unsafe queries. We address this challenge by introducing two comprehensive benchmarks: the Exaggerated Safety Benchmark…

计算与语言 · 计算机科学 2025-12-19 Shuzhou Yuan , Ercong Nie , Yinuo Sun , Chenxuan Zhao , William LaCroix , Michael Färber

The ability of large language models (LLMs) to follow user instructions is central to their reliability, safety, and usefulness. While prior studies assess instruction adherence in the model's main responses, we argue that it is also…

机器学习 · 计算机科学 2025-10-20 Yongchan Kwon , Shang Zhu , Federico Bianchi , Kaitlyn Zhou , James Zou

Large language models (LLMs) have demonstrated significant advancements in error handling. Current error-handling works are performed in a passive manner, with explicit error-handling instructions. However, in real-world scenarios, explicit…

计算与语言 · 计算机科学 2025-06-03 Jiayi Zeng , Yizhe Feng , Mengliang He , Wenhui Lei , Wei Zhang , Zeming Liu , Xiaoming Shi , Aimin Zhou

Given the recent impressive accomplishments of language models (LMs) for code generation, we explore the use of LMs as adaptive mutation and crossover operators for an evolutionary neural architecture search (NAS) algorithm. While NAS still…

神经与进化计算 · 计算机科学 2023-11-17 Angelica Chen , David M. Dohan , David R. So

Safety-aligned large language models (LLMs) often generate refusal responses to harmless queries due to the over-refusal problem. However, existing methods for mitigating over-refusal cannot maintain a low refusal ratio for harmless queries…

计算与语言 · 计算机科学 2026-04-21 Yupeng Qi , Ziyu Lyu , Lixin Cui , Lu Bai , Feng Xia

Large Language Models remain vulnerable to adversarial prompts that elicit toxic content even after safety alignment. We present ToxSearch, a black-box evolutionary framework that tests model safety by evolving prompts in a synchronous…

神经与进化计算 · 计算机科学 2026-01-27 Onkar Shelar , Travis Desell

Large language models are highly sensitive to prompts, but this sensitivity is usually studied through task-relevant instructions, demonstrations, or reasoning cues. In this paper, we study a different form of prompt sensitivity: whether…

计算与语言 · 计算机科学 2026-05-29 Pawel Batorski , Abtin Pourhadi , Jerzy Sarosiek , Przemyslaw Spurek , Paul Swoboda

Large language models (LLMs) have advanced to encompass extensive knowledge across diverse domains. Yet controlling what a large language model should not know is important for ensuring alignment and thus safe use. However, accurately and…

计算与语言 · 计算机科学 2024-11-01 Chris Yuhao Liu , Yaxuan Wang , Jeffrey Flanigan , Yang Liu

Self-evolving large language models (LLMs) learn by generating their own training tasks and solutions, reducing reliance on human-curated supervision. However, in many reasoning domains, the model must also validate generated tasks and…

人工智能 · 计算机科学 2026-05-28 Bowen Wei , Nan Wang , Yuqing Zhou , Jinhao Pan , Ziwei Zhu

Large Language Models (LLMs) generating unsafe responses to toxic prompts is a significant issue in their applications. While various efforts aim to address this safety concern, previous approaches often demand substantial human data…

计算与语言 · 计算机科学 2024-12-12 Yuxiao Lu , Arunesh Sinha , Pradeep Varakantham

Evolutionary algorithms have been successful in solving multi-objective optimization problems (MOPs). However, as a class of population-based search methodology, evolutionary algorithms require a large number of evaluations of the objective…

神经与进化计算 · 计算机科学 2024-08-16 Xueming Yan , Yaochu Jin

Fine-tuning language models is commonly believed to inevitably harm their safety, i.e., refusing to respond to harmful user requests, even when using harmless datasets, thus requiring additional safety measures. We challenge this belief…

机器学习 · 计算机科学 2025-08-19 Minseon Kim , Jin Myung Kwak , Lama Alssum , Bernard Ghanem , Philip Torr , David Krueger , Fazl Barez , Adel Bibi