AMaPO: Adaptive Margin-attached Preference Optimization for Language Model Alignment
Offline preference optimization offers a simpler and more stable alternative to RLHF for aligning language models. However, their effectiveness is critically dependent on ranking accuracy, a metric where further gains are highly impactful. This limitation arises from a fundamental problem that we identify and formalize as the Overfitting-Underfitting Dilemma: current margin designs cause models to apply excessive, wasteful gradients to correctly ranked samples (overfitting) while providing insufficient corrective signals for misranked ones (underfitting). To resolve this dilemma, we propose Adaptive Margin-attached Preference Optimization (AMaPO), a simple yet principled algorithm. AMaPO employs an instance-wise adaptive margin, refined by Z-normalization and exponential scaling, which dynamically reallocates learning effort by amplifying gradients for misranked samples and suppressing them for correct ones. Extensive experiments on widely used benchmarks demonstrate that AMaPO not only achieves better ranking accuracy and superior downstream alignment performance, but targeted analysis also confirms that it successfully mitigates the core overfitting and underfitting issues.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Adaptive Margin RLHF via Preference over Preferences
Margin-based optimization is fundamental to improving generalization and robustness in classification tasks. In the context of reward model learning from preferences within Reinforcement Learning from Human Feedback (RLH…
Reinforcement LearningMargin Adaptive DPO: Leveraging Reward Model for Granular Control in Preference Optimization
Direct Preference Optimization (DPO) has emerged as a simple and effective method for aligning large language models. However, its reliance on a fixed temperature parameter leads to suboptimal training on diverse prefere…
$α$-DPO: Adaptive Reward Margin is What Direct Preference Optimization Needs
Aligning large language models (LLMs) with human values and intentions is crucial for their utility, honesty, and safety. Reinforcement learning from human feedback (RLHF) is a popular approach to achieve this alignment,…
Computational EfficiencyCat-DPO: Category-Adaptive Safety Alignment
Aligning large language models with human preferences must balance two competing goals: responding helpfully to legitimate requests and reliably refusing harmful ones. Most preference-based safety alignment methods colla…
Diffusion-SDPO: Safeguarded Direct Preference Optimization for Diffusion Models
Text-to-image diffusion models deliver high-quality images, yet aligning them with human preferences remains challenging. We revisit diffusion-based Direct Preference Optimization (DPO) for these models and identify a cr…