English
Related papers

Related papers: PKU-SafeRLHF: Towards Multi-Level Safety Alignment…

200 papers

Reinforcement learning with human feedback (RLHF) fine-tunes a pretrained large language model (LLM) using user preference data, enabling it to generate content aligned with human preferences. However, due to privacy concerns, users may be…

Computation and Language · Computer Science 2025-04-10 Feijie Wu , Xiaoze Liu , Haoyu Wang , Xingchen Wang , Lu Su , Jing Gao

As large language models (LLMs) become increasingly integrated into daily applications, it is essential to ensure they operate fairly across diverse user demographics. In this work, we show that LLMs suffer from personalization bias, where…

Computation and Language · Computer Science 2025-02-12 Anvesh Rao Vijjini , Somnath Basu Roy Chowdhury , Snigdha Chaturvedi

Ensuring native-like quality of large language model (LLM) responses across many languages is challenging. To address this, we introduce MENLO, a framework that operationalizes the evaluation of native-like response quality based on…

Computation and Language · Computer Science 2026-03-03 Chenxi Whitehouse , Sebastian Ruder , Tony Lin , Oksana Kurylo , Haruka Takagi , Janice Lam , Nicolò Busetto , Denise Diaz , Francisco Guzmán

Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences, significantly enhancing the quality of interactions between humans and models. InstructGPT implements RLHF through…

Computation and Language · Computer Science 2023-10-10 Zheng Yuan , Hongyi Yuan , Chuanqi Tan , Wei Wang , Songfang Huang , Fei Huang

The remarkable abilities of large language models (LLMs) like GPT-4 partially stem from post-training processes like Reinforcement Learning from Human Feedback (RLHF) involving human preferences encoded in a reward model. However, these…

Artificial Intelligence · Computer Science 2023-12-06 Corby Rosset , Guoqing Zheng , Victor Dibia , Ahmed Awadallah , Paul Bennett

Large Language Models (LLMs) are widely used across sectors, yet their alignment with International Humanitarian Law (IHL) is not well understood. This study evaluates eight leading LLMs on their ability to refuse prompts that explicitly…

Computers and Society · Computer Science 2025-06-10 John Mavi , Diana Teodora Găitan , Sergio Coronado

Safety evaluations of large language models (LLMs) typically focus on universal risks like dangerous capabilities or undesirable propensities. However, millions use LLMs for personal advice on high-stakes topics like finance and health,…

Artificial Intelligence · Computer Science 2026-04-21 Manon Kempermann , Sai Suresh Macharla Vasu , Mahalakshmi Raveenthiran , Theo Farrell , Ingmar Weber

Humanitarian organizations face a critical choice: invest in costly commercial APIs or rely on free open-weight models for multilingual human rights monitoring. While commercial systems offer reliability, open-weight alternatives lack…

Computation and Language · Computer Science 2025-10-28 Poli Nemkova , Amrit Adhikari , Matthew Pearson , Vamsi Krishna Sadu , Mark V. Albert

Safety alignment approaches in large language models (LLMs) often lead to the over-refusal of benign queries, significantly diminishing their utility in sensitive scenarios. To address this challenge, we introduce FalseReject, a…

Computation and Language · Computer Science 2025-07-16 Zhehao Zhang , Weijie Xu , Fanyou Wu , Chandan K. Reddy

Children increasingly have access to Large Language Models (LLMs), which may expose them to responses that are developmentally inappropriate or require age-sensitive safety, guidance, and boundaries. Existing LLM safety evaluations largely…

Computation and Language · Computer Science 2026-05-27 Samee Arif , Angana Borah , Rada Mihalcea

Large language models (LLMs) are trained on a deluge of text data with limited quality control. As a result, LLMs can exhibit unintended or even harmful behaviours, such as leaking information, fake news or hate speech. Countermeasures,…

Machine Learning · Computer Science 2024-08-05 Jingtong Su , Julia Kempe , Karen Ullrich

We introduce a multi-turn benchmark for evaluating personalised alignment in LLM-based AI assistants, focusing on their ability to handle user-provided safety-critical contexts. Our assessment of ten leading models across five scenarios…

Human-Computer Interaction · Computer Science 2025-01-31 Lize Alberts , Benjamin Ellis , Andrei Lupu , Jakob Foerster

Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs). However, despite widespread adoption, the vast majority of work to-date has focused on first-class citizen…

Computation and Language · Computer Science 2024-07-04 John Dang , Arash Ahmadian , Kelly Marchisio , Julia Kreutzer , Ahmet Üstün , Sara Hooker

Aligning Large Language Model (LLM) responses with human preferences is vital for building safe and controllable AI systems. While preference optimization methods based on Plackett-Luce (PL) and Bradley-Terry (BT) models have shown promise,…

Artificial Intelligence · Computer Science 2026-03-23 Xiandong Zou , Wanyu Lin , Yuchen Li , Pan Zhou

Large Language Models (LLMs) rely on safety alignment to produce socially acceptable responses. However, this behavior is known to be brittle: further fine-tuning, even on benign or lightly contaminated data, can degrade safety and…

Machine Learning · Computer Science 2026-02-10 Kaustubh Ponkshe , Shaan Shah , Raghav Singhal , Praneeth Vepakomma

While there has been progress towards aligning Large Language Models (LLMs) with human values and ensuring safe behaviour at inference time, safety guards can easily be removed when fine tuned on unsafe and harmful datasets. While this…

Preference datasets are essential for incorporating human preferences into pre-trained language models, playing a key role in the success of Reinforcement Learning from Human Feedback. However, these datasets often demonstrate conflicting…

Artificial Intelligence · Computer Science 2025-10-23 Li Jiang , Yusen Wu , Junwu Xiong , Jingqing Ruan , Yichuan Ding , Qingpei Guo , Zujie Wen , Jun Zhou , Xiaotie Deng

Preference-based alignment like Reinforcement Learning from Human Feedback (RLHF) learns from pairwise preferences, yet the labels are often noisy and inconsistent. Existing uncertainty-aware approaches weight preferences, but ignore a more…

Machine Learning · Computer Science 2026-01-27 Tiejin Chen , Xiaoou Liu , Vishnu Nandam , Kuan-Ru Liou , Hua Wei

Safety evaluation of Large Language Models (LLMs) has largely focused on high-resource languages, leaving low-resource languages critically underserved. We present AlbanianLLMSafety, the first publicly available safety evaluation dataset…

Computation and Language · Computer Science 2026-05-29 Wajdi Zaghouani , Kholoud K. Aldous , Isra Fejzullaj

Large Reasoning Models (LRMs) have become powerful tools for complex problem solving, but their structured reasoning pathways can lead to unsafe outputs when exposed to harmful prompts. Existing safety alignment methods reduce harmful…

Artificial Intelligence · Computer Science 2025-10-24 Wonje Jeung , Sangyeon Yoon , Minsuk Kahng , Albert No