Boosting Reward Model with Preference-Conditional Multi-Aspect Synthetic Data Generation

• URL: http://arxiv.org/abs/2407.16008v1
• Date: Mon, 22 Jul 2024 19:21:55 GMT
• Title: Boosting Reward Model with Preference-Conditional Multi-Aspect Synthetic Data Generation
• Authors: Jiaming Shen, Ran Xu, Yennie Jun, Zhen Qin, Tianqi Liu, Carl Yang, Yi Liang, Simon Baumgartner, Michael Bendersky,
• Abstract summary: RMBoost is a novel synthetic preference data generation paradigm. It reduces labeling noise since preference pairs are constructed intentionally. It significantly boosts the performance of four distinct reward models.
• Score: 62.9933120822879
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Reward models (RMs) are crucial for aligning large language models (LLMs) with human preferences. They are trained using preference datasets where each example consists of one input prompt, two responses, and a preference label. As curating a high-quality human labeled preference dataset is both time-consuming and expensive, people often rely on existing powerful LLMs for preference label generation. This can potentially introduce noise and impede RM training. In this work, we present RMBoost, a novel synthetic preference data generation paradigm to boost reward model quality. Unlike traditional methods, which generate two responses before obtaining the preference label, RMBoost first generates one response and selects a preference label, followed by generating the second more (or less) preferred response conditioned on the pre-selected preference label and the first response. This approach offers two main advantages. First, RMBoost reduces labeling noise since preference pairs are constructed intentionally. Second, RMBoost facilitates the creation of more diverse responses by incorporating various quality aspects (e.g., helpfulness, relevance, completeness) into the prompts. We conduct extensive experiments across three diverse datasets and demonstrate that RMBoost outperforms other synthetic preference data generation techniques and significantly boosts the performance of four distinct reward models.

Related papers

• Anyprefer: An Agentic Framework for Preference Data Synthesis [62.3856754548222]
  We propose Anyprefer, a framework designed to synthesize high-quality preference data for aligning the target model. external tools are introduced to assist the judge model in accurately rewarding the target model's responses. The synthesized data is compiled into a new preference dataset, Anyprefer-V1, consisting of 58K high-quality preference pairs.
  arXiv  Detail & Related papers  (2025-04-27T15:21:59Z)
• More is Less: The Pitfalls of Multi-Model Synthetic Preference Data in DPO Safety Alignment [80.04449725137177]
  Direct Preference Optimization (DPO) has emerged as a simple, yet effective alternative to reinforcement learning from human feedback. Our study reveals a striking, safety-specific phenomenon associated with DPO alignment. Using solely self-generated responses for both chosen and rejected pairs significantly outperforms configurations that incorporate responses from stronger models.
  arXiv  Detail & Related papers  (2025-04-03T00:36:40Z)
• Rethinking Diverse Human Preference Learning through Principal Component Analysis [22.123631189289963]
  We introduce Decomposed Reward Models (DRMs), a novel approach that extracts diverse human preferences from binary comparisons. Our key insight is to represent human preferences as vectors and analyze them using Principal Component Analysis (PCA) DRMs effectively extract meaningful preference dimensions (e.g., helpfulness, safety, humor) and adapt to new users without additional training.
  arXiv  Detail & Related papers  (2025-02-18T18:55:26Z)
• Acoustic Model Optimization over Multiple Data Sources: Merging and Valuation [13.009945735929445]
  We propose a novel paradigm to solve salient problems plaguing the Automatic Speech Recognition field. In the first stage, multiple acoustic models are trained based upon different subsets of the complete speech data. In the second stage, two novel algorithms are utilized to generate a high-quality acoustic model.
  arXiv  Detail & Related papers  (2024-10-21T03:48:23Z)
• Reward-Augmented Data Enhances Direct Preference Alignment of LLMs [63.32585910975191]
  We introduce reward-conditioned Large Language Models (LLMs) that learn from the entire spectrum of response quality within the dataset. We show that our approach consistently boosts DPO by a considerable margin. Our method not only maximizes the utility of preference data but also mitigates the issue of unlearning, demonstrating its broad effectiveness beyond mere data expansion.
  arXiv  Detail & Related papers  (2024-10-10T16:01:51Z)
• Self-Boosting Large Language Models with Synthetic Preference Data [97.94185115047999]
  We introduce SynPO, a self-boosting paradigm that leverages synthetic preference data for model alignment. After four SynPO iterations, Llama3-8B and Mistral-7B show significant enhancements in instruction-following abilities. SynPO improves the general performance of LLMs on various tasks, validated by a 3.2 to 5.0 average score increase on the well-recognized Open LLM leaderboard.
  arXiv  Detail & Related papers  (2024-10-09T14:57:31Z)
• General Preference Modeling with Preference Representations for Aligning Language Models [51.14207112118503]
  We introduce preference representation learning, an approach that embeds responses into a latent space to capture intricate preference structures efficiently. We also propose preference score-based General Preference Optimization (GPO), which generalizes reward-based reinforcement learning from human feedback. Our method may enhance the alignment of foundation models with nuanced human values.
  arXiv  Detail & Related papers  (2024-10-03T04:22:55Z)
• REAL: Response Embedding-based Alignment for LLMs [1.9513983244114355]
  We propose a strategy for sampling a high-quality training dataset that focuses on acquiring the most informative response pairs. Experimental results indicate that choosing dissimilar response pairs enhances the direct alignment of LLMs. Our findings suggest that focusing on less similar pairs can improve the efficiency of LLM alignment, saving up to 65% of annotators' work.
  arXiv  Detail & Related papers  (2024-09-17T22:40:54Z)
• AEMLO: AutoEncoder-Guided Multi-Label Oversampling [6.255095509216069]
  AEMLO is an AutoEncoder-guided Oversampling technique for imbalanced multi-label data. We show that AEMLO outperforms the existing state-of-the-art methods with extensive empirical studies.
  arXiv  Detail & Related papers  (2024-08-23T14:01:33Z)
• Towards Comprehensive Preference Data Collection for Reward Modeling [15.495910034714187]
  Reinforcement Learning from Human Feedback (RLHF) facilitates the alignment of large language models with human preferences. We propose a framework for preference data collection, decomposing the process into four incremental steps. This structured approach ensures the collection of high-quality preferences while reducing reliance on human labor.
  arXiv  Detail & Related papers  (2024-06-24T09:40:39Z)
• Aligning Large Language Models with Self-generated Preference Data [72.99676237703099]
  We propose a new framework that boosts the alignment of large language models (LLMs) with human preferences. Our key idea is leveraging the human prior knowledge within the small (seed) data. We introduce a noise-aware preference learning algorithm to mitigate the risk of low quality within generated preference data.
  arXiv  Detail & Related papers  (2024-06-06T18:01:02Z)
• ROPO: Robust Preference Optimization for Large Language Models [59.10763211091664]
  We propose an iterative alignment approach that integrates noise-tolerance and filtering of noisy samples without the aid of external models. Experiments on three widely-used datasets with Mistral-7B and Llama-2-7B demonstrate that ROPO significantly outperforms existing preference alignment methods.
  arXiv  Detail & Related papers  (2024-04-05T13:58:51Z)
• Secrets of RLHF in Large Language Models Part II: Reward Modeling [134.97964938009588]
  We introduce a series of novel methods to mitigate the influence of incorrect and ambiguous preferences in the dataset. We also introduce contrastive learning to enhance the ability of reward models to distinguish between chosen and rejected responses.
  arXiv  Detail & Related papers  (2024-01-11T17:56:59Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.