SPARD: Self-Paced Curriculum for RL Alignment via Integrating Reward Dynamics and Data Utility
The evolution of Large Language Models (LLMs) is shifting the focus from single, verifiable tasks toward complex, open-ended real-world scenarios, imposing significant challenges on the post-training phase. In these settings, the scale and complexity of reward systems have grown significantly, transitioning toward multi-objective formulations that encompass a comprehensive spectrum of model capabilities and application contexts. However, traditional methods typically rely on fixed reward weights, ignoring non-stationary learning dynamics and struggling with data heterogeneity across dimensions. To address these issues, we propose SPARD, a framework that establishes an automated, self-paced curriculum by perceiving learning progress to dynamically adjust multi-objective reward weights and data importance, thereby synchronizing learning intent with data utility for optimal performance. Extensive experiments across multiple benchmarks demonstrate that SPARD significantly enhances model capabilities across all domains.
Code (0)
등록된 구현이 없습니다.
Similar Papers 제목 키워드 기반
Leveraging Self-Paced Curriculum Learning for Enhanced Modality Balance in Multimodal Conversational Emotion Recognition
Multimodal Emotion Recognition in Conversations (MERC) is a crucial task for understanding human interactions, where multimodal approaches integrating language, facial expressions, and vocal tone have achieved significan…
Multimodal Emotion RecognitionTraining with Growing Sets: A Simple Alternative to Curriculum Learning and Self Paced Learning
Curriculum learning and Self paced learning are popular topics in the machine learning that suggest to put the training samples in order by considering their difficulty levels. Studies in these topics show that starting …
Self Paced Gaussian Contextual Reinforcement Learning
Curriculum learning improves reinforcement learning (RL) efficiency by sequencing tasks from simple to complex. However, many self-paced curriculum methods rely on computationally expensive inner-loop optimizations, limi…
Reinforcement LearningReward-Machine-Guided, Self-Paced Reinforcement Learning
Self-paced reinforcement learning (RL) aims to improve the data efficiency of learning by automatically creating sequences, namely curricula, of probability distributions over contexts. However, existing techniques for s…
reinforcement-learningReinforcement LearningReinforcement Learning (RL)SPARD: Defending Harmful Fine-Tuning Attack via Safety Projection with Relevance-Diversity Data Selection
Fine-tuning large language models often undermines their safety alignment, a problem further amplified by harmful fine-tuning attacks in which adversarial data removes safeguards and induces unsafe behaviors. We propose …