English
Related papers

Related papers: Soft Best-of-n Sampling for Model Alignment

200 papers

Using natural language as a supervision for training visual recognition models holds great promise. Recent works have shown that if such supervision is used in the form of alignment between images and captions in large training datasets,…

Computer Vision and Pattern Recognition · Computer Science 2022-01-07 Ajinkya Tejankar , Maziar Sanjabi , Bichen Wu , Saining Xie , Madian Khabsa , Hamed Pirsiavash , Hamed Firooz

We present Distributional RewArds for Generative OptimizatioN (DRAGON), a versatile framework for fine-tuning media generation models towards a desired outcome. Compared with traditional reinforcement learning with human feedback (RLHF) or…

Sound · Computer Science 2025-11-18 Yatong Bai , Jonah Casebeer , Somayeh Sojoudi , Nicholas J. Bryan

Process Reward Models (PRMs) emerge as a promising approach for process supervision in mathematical reasoning of Large Language Models (LLMs), which aim to identify and mitigate intermediate errors in the reasoning processes. However, the…

Computation and Language · Computer Science 2025-06-06 Zhenru Zhang , Chujie Zheng , Yangzhen Wu , Beichen Zhang , Runji Lin , Bowen Yu , Dayiheng Liu , Jingren Zhou , Junyang Lin

Sampling is ubiquitous in machine learning methodologies. Due to the growth of large datasets and model complexity, we want to learn and adapt the sampling process while training a representation. Towards achieving this grand goal, a…

Machine Learning · Computer Science 2022-12-14 Jason Xiaotian Dou , Alvin Qingkai Pan , Runxue Bao , Haiyi Harry Mao , Lei Luo , Zhi-Hong Mao

Boosting is an extremely successful idea, allowing one to combine multiple low accuracy classifiers into a much more accurate voting classifier. In this work, we present a new and surprisingly simple Boosting algorithm that obtains a…

Machine Learning · Computer Science 2024-09-02 Mikael Møller Høgsgaard , Kasper Green Larsen , Markus Engelund Mathiasen

Recent work in large language modeling (LLMs) has used fine-tuning to align outputs with the preferences of a prototypical user. This work assumes that human preferences are static and homogeneous across individuals, so that aligning to a a…

Preference learning (PL) with large language models (LLMs) aims to align the LLMs' generations with human preferences. Previous work on reinforcement learning from human feedback (RLHF) has demonstrated promising results in in-distribution…

Machine Learning · Computer Science 2024-06-11 Chen Jia

Diffusion models achieve remarkable quality in image generation, but at a cost. Iterative denoising requires many time steps to produce high fidelity images. We argue that the denoising process is crucially limited by an accumulation of the…

Computer Vision and Pattern Recognition · Computer Science 2023-12-12 Hui Lu , Albert ali Salah , Ronald Poppe

Prompting has shown impressive success in enabling large pretrained language models (LMs) to perform diverse NLP tasks, especially when only few downstream data are available. Automatically finding the optimal prompt for each task, however,…

Computation and Language · Computer Science 2022-10-25 Mingkai Deng , Jianyu Wang , Cheng-Ping Hsieh , Yihan Wang , Han Guo , Tianmin Shu , Meng Song , Eric P. Xing , Zhiting Hu

Reinforcement learning from human feedback (RLHF) has emerged as an effective approach to aligning large language models (LLMs) to human preferences. RLHF contains three steps, i.e., human preference collecting, reward learning, and policy…

Computation and Language · Computer Science 2024-03-29 Hao Lang , Fei Huang , Yongbin Li

Aligning Large Language Models (LLMs) traditionally relies on costly training and human preference annotations. Self-alignment seeks to reduce these expenses by enabling models to align themselves. To further lower costs and achieve…

Computation and Language · Computer Science 2024-11-15 Somanshu Singla , Zhen Wang , Tianyang Liu , Abdullah Ashfaq , Zhiting Hu , Eric P. Xing

Optimising discrete data for a desired characteristic using gradient-based methods involves projecting the data into a continuous latent space and carrying out optimisation in this space. Carrying out global optimisation is difficult as…

Machine Learning · Computer Science 2019-05-27 Omar Mahmood , José Miguel Hernández-Lobato

Normalization methods are a central building block in the deep learning toolbox. They accelerate and stabilize training, while decreasing the dependence on manually tuned learning rate schedules. When learning from multi-modal…

Machine Learning · Computer Science 2018-10-15 Lucas Deecke , Iain Murray , Hakan Bilen

The dominant approach to generating from language models subject to some constraint is locally constrained decoding (LCD), incrementally sampling tokens at each time step such that the constraint is never violated. Typically, this is…

Rejection sampling methods have recently been proposed to improve the performance of discriminator-based generative models. However, these methods are only optimal under an unlimited sampling budget, and are usually applied to a generator…

Machine Learning · Computer Science 2024-03-04 Alexandre Verine , Muni Sreenivas Pydi , Benjamin Negrevergne , Yann Chevaleyre

Uncertainty sampling, a popular active learning algorithm, is used to reduce the amount of data required to learn a classifier, but it has been observed in practice to converge to different parameters depending on the initialization and…

Machine Learning · Computer Science 2018-12-06 Stephen Mussmann , Percy Liang

In aligning large language models (LLMs), reward models have played an important role, but are standardly trained as discriminative models and rely only on labeled human preference data. In this paper, we explore methods that train reward…

Computation and Language · Computer Science 2026-01-27 Chenglong Wang , Yang Gan , Yifu Huo , Yongyu Mu , Qiaozhi He , Murun Yang , Bei Li , Tong Xiao , Chunliang Zhang , Tongran Liu , Jingbo Zhu

Decoding methods for large language models often trade-off between diversity of outputs and parallelism of computation. Methods such as beam search and Gumbel top-k sampling can guarantee a different output for each element of the beam, but…

Computation and Language · Computer Science 2023-06-02 Luke Vilnis , Yury Zemlyanskiy , Patrick Murray , Alexandre Passos , Sumit Sanghai

Large language models (LLMs) have revolutionized the role of AI, yet pose potential social risks. To steer LLMs towards human preference, alignment technologies have been introduced and gained increasing attention. Nevertheless, existing…

Computation and Language · Computer Science 2024-10-01 Shitong Duan , Xiaoyuan Yi , Peng Zhang , Yan Liu , Zheng Liu , Tun Lu , Xing Xie , Ning Gu

Finetuning language models with reinforcement learning (RL), e.g. from human feedback (HF), is a prominent method for alignment. But optimizing against a reward model can improve on reward while degrading performance in other areas, a…

Computation and Language · Computer Science 2023-12-14 Michael Noukhovitch , Samuel Lavoie , Florian Strub , Aaron Courville