English
Related papers

Related papers: FlipGuard: Defending Preference Alignment against …

200 papers

Recently, tremendous strides have been made to align the generation of Large Language Models (LLMs) with human values to mitigate toxic or unhelpful content. Leveraging Reinforcement Learning from Human Feedback (RLHF) proves effective and…

Computation and Language · Computer Science 2024-06-05 Mingye Zhu , Yi Liu , Lei Zhang , Junbo Guo , Zhendong Mao

Aligning the output of Large Language Models (LLMs) with human preferences (e.g., by means of reinforcement learning with human feedback, or RLHF) is essential for ensuring their effectiveness in real-world scenarios. Despite significant…

Artificial Intelligence · Computer Science 2024-10-23 Pietro Bernardelle , Gianluca Demartini

Reinforcement learning has been explored for many problems, from video games with deterministic environments to portfolio and operations management in which scenarios are stochastic; however, there have been few attempts to test these…

General Finance · Quantitative Finance 2024-02-19 Sherly Alfonso-Sánchez , Jesús Solano , Alejandro Correa-Bahnsen , Kristina P. Sendova , Cristián Bravo

Stochastic gradient descent (SGD) is a premium optimization method for training neural networks, especially for learning objectively defined labels such as image objects and events. When a neural network is instead faced with subjectively…

Neural and Evolutionary Computing · Computer Science 2022-04-15 Kosmas Pinitas , Konstantinos Makantasis , Antonios Liapis , Georgios N. Yannakakis

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

Distributed training and increasing the gradient update frequency are practical strategies to accelerate learning and improve performance, but both exacerbate a central challenge: \textit{policy lag}, which is the mismatch between the…

Large Language Models (LLMs) are often aligned using contrastive alignment objectives and preference pair datasets. The interaction between model, paired data, and objective makes alignment a complicated procedure, sometimes producing…

Backward error analysis allows finding a modified loss function, which the parameter updates really follow under the influence of an optimization method. The additional loss terms included in this modified function is called implicit…

Machine Learning · Computer Science 2025-03-06 Jinwoo Lim , Suhyun Kim , Soo-Mook Moon

Language models serve as proxies for human preference judgements in alignment and evaluation, yet they exhibit systematic miscalibration, prioritizing superficial patterns over substantive qualities. This bias manifests as overreliance on…

Computation and Language · Computer Science 2026-03-05 Anirudh Bharadwaj , Chaitanya Malaviya , Nitish Joshi , Mark Yatskar

Beyond simple text generation, Large Language Models (LLMs) have evolved into agentic systems capable of planning and interacting with external tools to solve complex tasks. This evolution involves fine-tuning LLMs on agent-specific tasks…

Computation and Language · Computer Science 2025-11-18 Dongyoon Hahm , Taywon Min , Woogyeol Jin , Kimin Lee

Fine-tuning large pre-trained language models for downstream tasks remains a critical challenge in natural language processing. This paper presents an empirical analysis comparing two efficient fine-tuning methods - BitFit and adapter…

Machine Learning · Computer Science 2024-01-09 Nigel Doering , Cyril Gorlla , Trevor Tuttle , Adhvaith Vijay

Large-scale generative models have shown impressive image-generation capabilities, propelled by massive data. However, this often inadvertently leads to the generation of harmful or inappropriate content and raises copyright concerns.…

Machine Learning · Computer Science 2025-03-11 Myeongseob Ko , Henry Li , Zhun Wang , Jonathan Patsenker , Jiachen T. Wang , Qinbin Li , Ming Jin , Dawn Song , Ruoxi Jia

Preference-based reinforcement learning (PbRL) provides a natural way to align RL agents' behavior with human desired outcomes, but is often restrained by costly human feedback. To improve feedback efficiency, most existing PbRL methods…

Machine Learning · Computer Science 2024-07-08 Xiao Hu , Jianxiong Li , Xianyuan Zhan , Qing-Shan Jia , Ya-Qin Zhang

Alignment via reinforcement learning from human feedback (RLHF) has become the dominant paradigm for controlling the quality of outputs from large language models (LLMs). However, existing theories do not provide strong justification for…

Machine Learning · Computer Science 2026-05-19 Jihun Yun , Juno Kim , Jongho Park , Junhyuck Kim , Jongha Jon Ryu , Jaewoong Cho , Kwang-Sung Jun

Robustness evaluation against adversarial examples has become increasingly important to unveil the trustworthiness of the prevailing deep models in natural language processing (NLP). However, in contrast to the computer vision domain where…

Computation and Language · Computer Science 2022-12-20 Bairu Hou , Jinghan Jia , Yihua Zhang , Guanhua Zhang , Yang Zhang , Sijia Liu , Shiyu Chang

In this article, we investigate the alignment of Large Language Models according to human preferences. We discuss the features of training a Preference Model, which simulates human preferences, and the methods and details we found essential…

Machine Learning · Computer Science 2024-10-03 Alexey Kutalev , Sergei Markoff

Reinforcement Learning from Human Feedback (RLHF) aligns language models to human preferences by employing a singular reward model derived from preference data. However, such an approach overlooks the rich diversity of human preferences…

Computation and Language · Computer Science 2024-12-30 Souradip Chakraborty , Jiahao Qiu , Hui Yuan , Alec Koppel , Furong Huang , Dinesh Manocha , Amrit Singh Bedi , Mengdi Wang

Large Language Models (LLMs) have emerged as promising solutions for a variety of medical and clinical decision support applications. However, LLMs are often subject to different types of biases, which can lead to unfair treatment of…

Computation and Language · Computer Science 2024-08-23 Raphael Poulain , Hamed Fayyaz , Rahmatollah Beheshti

Collecting high-quality preference datasets for reinforcement learning from human feedback (RLHF) is resource-intensive and challenging. As a result, researchers often train reward models on extensive offline datasets which aggregate…

Machine Learning · Computer Science 2024-12-17 Shambhavi Krishna , Aishwarya Sahoo

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…