English
Related papers

Related papers: On Monotonicity in AI Alignment

200 papers

Offline preference optimization methods are efficient for large language models (LLMs) alignment. Direct Preference optimization (DPO)-like learning, one of the most popular approaches, stands out for its efficiency in reward modeling.…

Machine Learning · Computer Science 2026-05-26 Xiaobo Wang , Zixia Jia , Jiaqi Li , Qi Liu , Zilong Zheng

Direct Alignment Algorithms (DAAs), such as Direct Preference Optimisation (DPO) and Identity Preference Optimisation (IPO), have emerged as alternatives to online Reinforcement Learning from Human Feedback (RLHF) algorithms such as…

Computation and Language · Computer Science 2024-10-21 Zhengyan Shi , Sander Land , Acyr Locatelli , Matthieu Geist , Max Bartolo

The dominant practice of AI alignment assumes (1) that preferences are an adequate representation of human values, (2) that human rationality can be understood in terms of maximizing the satisfaction of preferences, and (3) that AI systems…

Artificial Intelligence · Computer Science 2024-11-12 Tan Zhi-Xuan , Micah Carroll , Matija Franklin , Hal Ashton

Synthesizing realistic and diverse anomalous samples from limited data is vital for robust model generalization. However, existing methods struggle to reconcile fidelity and diversity, often hampered by distribution misalignment and…

Computer Vision and Pattern Recognition · Computer Science 2026-05-19 Fuyun Wang , Yuanzhi Wang , Xu Guo , Sujia Huang , Tong Zhang , Dan Wang , Hui Yan , Xin Liu , Zhen Cui

Preference learning algorithms (e.g., RLHF and DPO) are frequently used to steer LLMs to produce generations that are more preferred by humans, but our understanding of their inner workings is still limited. In this work, we study the…

Machine Learning · Computer Science 2024-11-01 Angelica Chen , Sadhika Malladi , Lily H. Zhang , Xinyi Chen , Qiuyi Zhang , Rajesh Ranganath , Kyunghyun Cho

Learning performance can show non-monotonic behavior. That is, more data does not necessarily lead to better models, even on average. We propose three algorithms that take a supervised learning model and make it perform more monotone. We…

Machine Learning · Computer Science 2019-11-26 Tom J. Viering , Alexander Mey , Marco Loog

Direct preference optimization (\texttt{DPO}) has emerged as a promising approach for solving the alignment problem in AI. In this paper, we make two counter-intuitive observations about \texttt{DPO}. First, we show that \texttt{DPO} loss…

Preference based alignment objectives implicitly assume that all human preferences are expressed with equal strength. In practice, however, preference strength varies across individuals and contexts -- a phenomenon established in behavioral…

Machine Learning · Computer Science 2026-01-13 Saki Imai , Pedram Heydari , Anthony Sicilia , Asteria Kaeberlein , Katherine Atwell , Malihe Alikhani

Aligning large language models (LLMs) with human intent is critical for enhancing their performance across a variety of tasks. Standard alignment techniques, such as Direct Preference Optimization (DPO), often rely on the binary…

Computation and Language · Computer Science 2025-04-01 Yuxiang Guo , Lu Yin , Bo Jiang , Jiaqi Zhang

Efficient preference optimization algorithms such as Direct Preference Optimization (DPO) have become a popular approach in aligning large language models (LLMs) with human preferences. These algorithms implicitly treat the LLM as a reward…

Computation and Language · Computer Science 2025-07-29 Tong Liu , Xiao Yu , Wenxuan Zhou , Jindong Gu , Volker Tresp

Learning monotonic models with respect to a subset of the inputs is a desirable feature to effectively address the fairness, interpretability, and generalization issues in practice. Existing methods for learning monotonic neural networks…

Machine Learning · Computer Science 2022-12-16 Xingchao Liu , Xing Han , Na Zhang , Qiang Liu

Compound AI systems, comprising multiple interacting components such as LLMs, foundation models, and external tools, have demonstrated remarkable improvements compared to single models in various tasks. To ensure their effective deployment…

Machine Learning · Computer Science 2026-03-09 Xiangwen Wang , Yibo Jacky Zhang , Zhoujie Ding , Katherine Tsai , Haolun Wu , Sanmi Koyejo

Evaluating the pedagogical quality of AI tutors remains challenging: standard NLG metrics do not determine whether responses identify mistakes, scaffold reasoning, or avoid revealing the answers. For the task of mistake remediation, we…

Computation and Language · Computer Science 2026-03-26 Kseniia Petukhova , Ekaterina Kochmar

Aligning large language models (LLMs) with human preferences remains a key challenge in AI. Preference-based optimization methods, such as Reinforcement Learning with Human Feedback (RLHF) and Direct Preference Optimization (DPO), rely on…

Computation and Language · Computer Science 2025-05-26 Xuan Qi , Jiahao Qiu , Xinzhe Juan , Yue Wu , Mengdi Wang

We study the post-training of large language models (LLMs) with human preference data. Recently, direct preference optimization and its variants have shown considerable promise in aligning language models, eliminating the need for reward…

Machine Learning · Computer Science 2025-05-14 Teng Xiao , Zhen Ge , Sujay Sanghavi , Tian Wang , Julian Katz-Samuels , Marc Versage , Qingjun Cui , Trishul Chilimbi

Preference learning has become the foundation of aligning Large Language Models (LLMs) with human intent. Popular methods, such as Direct Preference Optimization (DPO), minimize surrogate losses as proxies for the intractable pairwise…

Machine Learning · Computer Science 2026-05-01 Mehryar Mohri , Yutao Zhong

Preference alignment in Large Language Models (LLMs) has significantly improved their ability to adhere to human instructions and intentions. However, existing direct alignment algorithms primarily focus on relative preferences and often…

Machine Learning · Computer Science 2025-05-13 Shenao Zhang , Zhihan Liu , Boyi Liu , Yufeng Zhang , Yingxiang Yang , Yongfei Liu , Liyu Chen , Tao Sun , Zhaoran Wang

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

Direct Preference Optimization (DPO) has emerged as a predominant alignment method for diffusion models, facilitating off-policy training without explicit reward modeling. However, its reliance on large-scale, high-quality human preference…

Computer Vision and Pattern Recognition · Computer Science 2026-02-09 Khiem Pham , Quang Nguyen , Tung Nguyen , Jingsen Zhu , Michele Santacatterina , Dimitris Metaxas , Ramin Zabih

Aligning large language models (LLMs) with diverse and multifaceted user preferences is a fundamental challenge in personalized AI systems. Existing multi-objective alignment methods either rely on costly training or require pre-trained…

Computation and Language · Computer Science 2026-05-26 Linhao Luo , Thuy-Trang Vu , Van-Anh Nguyen , Junae Kim , Gholamreza Haffari , Dinh Phung