English
Related papers

Related papers: Not All Preferences Are Created Equal: Stability-A…

200 papers

Preference optimization methods such as DPO align large language models (LLMs) using paired comparisons, but their effectiveness can be highly sensitive to the quality and difficulty of preference pairs. A common heuristic treats…

Artificial Intelligence · Computer Science 2026-02-03 Jinlong Pang , Zhaowei Zhu , Na Di , Yichi Zhang , Yaxuan Wang , Chen Qian , Yang Liu

In neural network models, soft mixtures of fixed candidate components (e.g., logic gates and sub-networks) are often used during training for stable optimization, while hard selection is typically used at inference. This raises questions…

Machine Learning · Computer Science 2026-03-17 Youngsung Kim

We study the performance of stochastic first-order methods for finding saddle points of convex-concave functions. A notorious challenge faced by such methods is that the gradients can grow arbitrarily large during optimization, which may…

Machine Learning · Computer Science 2024-06-10 Gergely Neu , Nneka Okolo

Preference-based reinforcement learning (RL) is a key paradigm for aligning policies with human judgments, yet its theoretical behavior in distributed settings where preference data are fragmented across heterogeneous users remains poorly…

Machine Learning · Computer Science 2026-05-21 Zhanhong Jiang

Aligning large language models (LLMs) with human preferences typically demands vast amounts of meticulously curated data, which is both expensive and prone to labeling noise. We propose Stackelberg Game Preference Optimization (SGPO), a…

Machine Learning · Computer Science 2026-01-22 Xu Chu , Zhixin Zhang , Tianyu Jia , Yujie Jin

Information-based data selection for instruction tuning is compelling: maximizing the log-determinant of the Fisher information yields a monotone submodular objective, enabling greedy algorithms to achieve a $(1-1/e)$ approximation under a…

Direct Preference Optimization (DPO) is broadly utilized for aligning Large Language Models (LLMs) with human values because of its flexibility. Despite its effectiveness, it has been observed that the capability of DPO to generate…

Machine Learning · Computer Science 2025-05-20 Wenqiao Zhu , Ji Liu , Lulu Wang , Jun Wu , Yulun Zhang

Alignment of large language models (LLMs) typically involves training a reward model on preference data, followed by policy optimization with respect to the reward model. However, optimizing policies with respect to a single reward model…

Machine Learning · Computer Science 2025-07-23 Debangshu Banerjee , Kintan Saha , Aditya Gopalan

Preference-based reinforcement learning offers a scalable alternative to manual reward engineering by learning reward structures from comparative feedback. However, large-scale preference datasets, whether collected from crowdsourced…

Robotics · Computer Science 2026-05-04 Ziqin Yuan , Ruiqi Wang , Dezhong Zhao , Baijian Yang , Byung-Cheol Min

Stable matching is a fundamental problem studied both in economics and computer science. The task is to find a matching between two sides of agents that have preferences over who they want to be matched with. A matching is stable if no pair…

Computer Science and Game Theory · Computer Science 2024-03-11 Juho Hirvonen , Sara Ranjbaran

Reinforcement Learning from Human Feedback (RLHF) has become central to aligning large language models with human values, typically by first learning a reward model from preference data which is then used to update the model with…

Machine Learning · Computer Science 2025-10-21 Keertana Chidambaram , Karthik Vinay Seetharaman , Vasilis Syrgkanis

The vision of an inclusive World Wide Web is impeded by a severe linguistic divide, particularly for communities in low-resource regions of Southeast Asia. While large language models (LLMs) offer a potential solution for translation, their…

Computation and Language · Computer Science 2026-03-23 Zhixiang Lu , Chong Zhang , Yulong Li , Angelos Stefanidis , Anh Nguyen , Imran Razzak , Jionglong Su , Zhengyong Jiang

Data augmentation has been widely employed to improve the generalization of deep neural networks. Most existing methods apply fixed or random transformations. However, we find that sample difficulty evolves along with the model's…

Machine Learning · Computer Science 2025-10-02 Suorong Yang , Jie Zong , Lihang Wang , Ziheng Qin , Hai Gan , Pengfei Zhou , Kai Wang , Yang You , Furao Shen

Group-based policy optimization methods like GRPO and GSPO have become standard for training multimodal models, leveraging group-wise rollouts and relative advantage estimation. However, they suffer from a critical \emph{gradient vanishing}…

Machine Learning · Computer Science 2025-11-25 Zengjie Hu , Jiantao Qiu , Tianyi Bai , Haojin Yang , Binhang Yuan , Qi Jing , Conghui He , Wentao Zhang

Distributed stochastic gradient descent (SGD) has attracted considerable recent attention due to its potential for scaling computational resources, reducing training time, and helping protect user privacy in machine learning. However, the…

Machine Learning · Computer Science 2025-02-27 Siyuan Yu , Wei Chen , H. Vincent Poor

Agentic reinforcement learning (ARL) has rapidly gained attention as a promising paradigm for training agents to solve complex, multi-step interactive tasks. Despite encouraging early results, ARL remains highly unstable, often leading to…

This paper addresses the problem of preference learning, which aims to align robot behaviors through learning user specific preferences (e.g. "good pull-over location") from visual demonstrations. Despite its similarity to learning factual…

Robotics · Computer Science 2025-01-16 Sadanand Modak , Noah Patton , Isil Dillig , Joydeep Biswas

Alignment with human preferences is commonly framed using a universal reward function, even though human preferences are inherently heterogeneous. We formalize this heterogeneity by introducing user types and examine the limits of the…

Artificial Intelligence · Computer Science 2025-02-25 Ali Shirali , Arash Nasr-Esfahany , Abdullah Alomar , Parsa Mirtaheri , Rediet Abebe , Ariel Procaccia

Efficiently aligning large-scale video diffusion models with human intent requires a scalable and trajectory-aware pathway that bridges the inherent discrepancy between training noise distributions and practical inference trajectories.…

Computer Vision and Pattern Recognition · Computer Science 2026-05-11 Jingyuan Zhu , Biaolong Chen , Le Zhang , Aixi Zhang , Hao Jiang , Pipei Huang

Reinforcement Learning from Human Feedback (RLHF) has become central to aligning large language models with human values, typically by first learning a reward model from preference data which is then used to update the model with…

Artificial Intelligence · Computer Science 2025-10-20 Keertana Chidambaram , Karthik Vinary Seetharaman , Vasilis Syrgkanis