English
Related papers

Related papers: Policy Teaching via Data Poisoning in Learning fro…

200 papers

Reinforcement Learning from Human Feedback (RLHF) is a popular method for aligning Language Models (LM) with human values and preferences. RLHF requires a large number of preference pairs as training data, which are often used in both the…

Computation and Language · Computer Science 2024-08-07 Tim Baumgärtner , Yang Gao , Dana Alon , Donald Metzler

The prevalent deployment of learning from human preferences through reinforcement learning (RLHF) relies on two important approximations: the first assumes that pairwise preferences can be substituted with pointwise rewards. The second…

Artificial Intelligence · Computer Science 2023-11-23 Mohammad Gheshlaghi Azar , Mark Rowland , Bilal Piot , Daniel Guo , Daniele Calandriello , Michal Valko , Rémi Munos

Learning reward models from pairwise comparisons is a fundamental component in a number of domains, including autonomous control, conversational agents, and recommendation systems, as part of a broad goal of aligning automated decisions…

Machine Learning · Computer Science 2024-10-10 Junlin Wu , Jiongxiao Wang , Chaowei Xiao , Chenguang Wang , Ning Zhang , Yevgeniy Vorobeychik

This work tackles the problem of overoptimization in reinforcement learning from human feedback (RLHF), a prevalent technique for aligning models with human preferences. RLHF relies on reward or preference models trained on \emph{fixed…

Machine Learning · Computer Science 2025-03-11 Dhawal Gupta , Adam Fisch , Christoph Dann , Alekh Agarwal

Data poisoning attacks aim to manipulate the model produced by a learning algorithm by adversarially modifying the training set. We consider differential privacy as a defensive measure against this type of attack. We show that such learners…

Machine Learning · Computer Science 2019-07-08 Yuzhe Ma , Xiaojin Zhu , Justin Hsu

Recent advancements in Reinforcement Learning with Human Feedback (RLHF) have significantly impacted the alignment of Large Language Models (LLMs). The sensitivity of reinforcement learning algorithms such as Proximal Policy Optimization…

Machine Learning · Computer Science 2025-06-10 Pankayaraj Pathmanathan , Souradip Chakraborty , Xiangyu Liu , Yongyuan Liang , Furong Huang

In this paper, we take a step towards a deeper understanding of learning from human preferences by systematically comparing the paradigm of reinforcement learning from human feedback (RLHF) with the recently proposed paradigm of direct…

Machine Learning · Computer Science 2024-06-06 Andi Nika , Debmalya Mandal , Parameswaran Kamalaruban , Georgios Tzannetos , Goran Radanović , Adish Singla

Direct preference optimization (DPO) is a form of reinforcement learning from human feedback (RLHF) where the policy is learned directly from preferential feedback. Although many models of human preferences exist, the critical task of…

Machine Learning · Computer Science 2025-03-04 Branislav Kveton , Xintong Li , Julian McAuley , Ryan Rossi , Jingbo Shang , Junda Wu , Tong Yu

We study a security threat to reinforcement learning where an attacker poisons the learning environment to force the agent into executing a target policy chosen by the attacker. As a victim, we consider RL agents whose objective is to find…

Machine Learning · Computer Science 2020-08-20 Amin Rakhsha , Goran Radanovic , Rati Devidze , Xiaojin Zhu , Adish Singla

Data poisoning is a type of adversarial attack on training data where an attacker manipulates a fraction of data to degrade the performance of machine learning model. Therefore, applications that rely on external data-sources for training…

Machine Learning · Computer Science 2021-04-28 Sanjay Seetharaman , Shubham Malaviya , Rosni KV , Manish Shukla , Sachin Lodha

Offline Reinforcement Learning from Human Feedback (RLHF) pipelines such as Direct Preference Optimization (DPO) train on a pre-collected preference dataset, which makes them vulnerable to preference poisoning attack. We study label flip…

Machine Learning · Computer Science 2026-05-26 Chenye Yang , Weiyu Xu , Lifeng Lai

To design rewards that align with human goals, Reinforcement Learning from Human Feedback (RLHF) has emerged as a prominent technique for learning reward functions from human preferences and optimizing policies via reinforcement learning…

Machine Learning · Computer Science 2025-05-14 Taehyun Cho , Seokhun Ju , Seungyub Han , Dohyeong Kim , Kyungjae Lee , Jungwoo Lee

Data poisoning attacks -- where an adversary can modify a small fraction of training data, with the goal of forcing the trained classifier to high loss -- are an important threat for machine learning in many applications. While a body of…

Machine Learning · Computer Science 2020-02-21 Yizhen Wang , Somesh Jha , Kamalika Chaudhuri

We study a security threat to reinforcement learning where an attacker poisons the learning environment to force the agent into executing a target policy chosen by the attacker. As a victim, we consider RL agents whose objective is to find…

Machine Learning · Computer Science 2020-11-24 Amin Rakhsha , Goran Radanovic , Rati Devidze , Xiaojin Zhu , Adish Singla

Preference learning is a central component for aligning current LLMs, but this process can be vulnerable to data poisoning attacks. To address this concern, we introduce PoisonBench, a benchmark for evaluating large language models'…

Cryptography and Security · Computer Science 2025-06-09 Tingchen Fu , Mrinank Sharma , Philip Torr , Shay B. Cohen , David Krueger , Fazl Barez

Reinforcement Learning with Human Feedback (RLHF) is a methodology designed to align Large Language Models (LLMs) with human preferences, playing an important role in LLMs alignment. Despite its advantages, RLHF relies on human annotators…

Artificial Intelligence · Computer Science 2024-06-21 Jiongxiao Wang , Junlin Wu , Muhao Chen , Yevgeniy Vorobeychik , Chaowei Xiao

We consider data poisoning attacks, a class of adversarial attacks on machine learning where an adversary has the power to alter a small fraction of the training data in order to make the trained classifier satisfy certain objectives. While…

Machine Learning · Computer Science 2018-08-29 Yizhen Wang , Kamalika Chaudhuri

To improve human-preference alignment training, current research has developed numerous preference datasets consisting of preference pairs labeled as "preferred" or "dispreferred". These preference pairs are typically used to encode human…

Computation and Language · Computer Science 2024-10-08 Chenglong Wang , Yang Gan , Yifu Huo , Yongyu Mu , Qiaozhi He , Murun Yang , Tong Xiao , Chunliang Zhang , Tongran Liu , Jingbo Zhu

Reinforcement learning with human feedback (RLHF) has become the dominant method to align large models to user preferences. Unlike fine-tuning, for which there are many studies regarding training data memorization, it is not clear how…

Machine Learning · Computer Science 2024-10-28 Aneesh Pappu , Billy Porter , Ilia Shumailov , Jamie Hayes

Recommender systems play a crucial role in helping users to find their interested information in various web services such as Amazon, YouTube, and Google News. Various recommender systems, ranging from neighborhood-based,…

Cryptography and Security · Computer Science 2021-01-11 Hai Huang , Jiaming Mu , Neil Zhenqiang Gong , Qi Li , Bin Liu , Mingwei Xu
‹ Prev 1 2 3 10 Next ›