相关论文: Beyond Correctness: Evaluating Subjective Writing …

Reinforcement Learning from Human Feedback (RLHF) is a widely used framework for the training of language models. However, the process of using RLHF to develop a language model that is well-aligned presents challenges, especially when it…

Reward models are used throughout the post-training of language models to capture nuanced signals from preference data and provide a training target for optimization across instruction following, reasoning, safety, and more domains. The…

Reinforcement Learning from Human Feedback (RLHF) has become a crucial technology for aligning language models with human values and intentions, enabling models to produce more helpful and harmless responses. Reward models are trained as…

While Reinforcement Learning from Human Feedback (RLHF) is widely used to align Large Language Models (LLMs) with human preferences, it typically assumes homogeneous preferences across users, overlooking diverse human values and minority…

Reward models (RMs) are at the crux of successfully using RLHF to align pretrained models to human preferences, yet there has been relatively little study that focuses on evaluation of those models. Evaluating reward models presents an…

Reinforcement Learning from Human Feedback has become the standard paradigm for language model alignment, where reward models directly determine alignment effectiveness. In this work, we focus on how to evaluate the generalizability of…

Reinforcement Learning with Human Feedback (RLHF) has emerged as a key paradigm for task-specific fine-tuning of language models using human preference data. While numerous publicly available preference datasets provide pairwise comparisons…

Learning human preferences in language models remains fundamentally challenging, as reward modeling relies on subtle, subjective comparisons or shades of gray rather than clear-cut labels. This study investigates the limits of current…

Pluralistic alignment has emerged as a critical frontier in the development of Large Language Models (LLMs), with reward models (RMs) serving as a central mechanism for capturing diverse human values. While benchmarks for general response…

Recent advances in generative AI have been driven by alignment techniques such as reinforcement learning from human feedback (RLHF). RLHF and related techniques typically involve constructing a dataset of binary or ranked choice human…

Reinforcement Learning from Human Feedback (RLHF) is a powerful paradigm for aligning foundation models to human values and preferences. However, current RLHF techniques cannot account for the naturally occurring differences in individual…

Reward models are critical in techniques like Reinforcement Learning from Human Feedback (RLHF) and Inference Scaling Laws, where they guide language model alignment and select optimal responses. Despite their importance, existing reward…

Reward modeling is crucial for aligning large language models with human preferences, yet current approaches lack a principled mathematical framework for leveraging ordinal preference data. When human annotators provide graded preferences…

Large language models (LLMs) are increasingly deployed via public-facing interfaces to interact with millions of users, each with diverse preferences. Despite this, preference tuning of LLMs predominantly relies on reward models trained…

We introduce a new benchmark for reward models that quantifies their ability to produce strong language models through RLHF (Reinforcement Learning from Human Feedback). The gold-standard approach is to run a full RLHF training pipeline and…

Reinforcement Learning from Human Feedback (RLHF) assumes annotator preferences reflect stable internal states. We challenge this through three experiments spanning the preference pipeline. In a human choice blindness study, 91% of…

The goal of aligning language models to human preferences requires data that reveal these preferences. Ideally, time and money can be spent carefully collecting and tailoring bespoke preference data to each downstream application. However,…

Despite notable advancements in Multimodal Large Language Models (MLLMs), most state-of-the-art models have not undergone thorough alignment with human preferences. This gap exists because current alignment research has primarily achieved…

Faithfulness, expressiveness, and elegance is the constant pursuit in machine translation. However, traditional metrics like \textit{BLEU} do not strictly align with human preference of translation quality. In this paper, we explore…

Recent advances in aligning Large Language Models with human preferences have benefited from larger reward models and better preference data. However, most of these methodologies rely on the accuracy of the reward model. The reward models…