arXiv:2505. 10892v2 Announce Type: replace Abstract: Post-training LLMs with RLHF and preference optimization methods (e.
arXiv:2512. 15765v3 Announce Type: replace Abstract: Data valuation is a natural framework for understanding which preference datasets matter most when aligning a Large Language Model (LLM) using multiple sources.
arXiv:2506. 13702v4 Announce Type: replace-cross Abstract: Single-trajectory preference optimization methods learn from datasets of ((prompt, response, reward)) tuples, offering a practical alternative to pairwise preference learning by directly leveraging scalar feedback.
Aligning large language models with diverse and heterogeneous human values requires multi-objective alignment methods to effectively trade off conflicting preference dimensions. Current methods achieve this trade-off by training policies conditioned on preference vectors and leveraging online direct preference optimization.
arXiv:2510. 01167v2 Announce Type: replace-cross Abstract: Aligning large language models to human preferences is inherently multidimensional, yet most pipelines collapse heterogeneous signals into a single objective.
arXiv:2607. 29559v1 Announce Type: new Abstract: Reinforcement Learning (RL) systems are typically trained using a single, well-specified scalar reward function.
arXiv:2606. 04807v1 Announce Type: new Abstract: Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge: unlike verifiable tasks, bias lacks a single ground truth, creating a high-variance, subjective reward landscape.
arXiv:2602. 07764v2 Announce Type: replace-cross Abstract: Multi-objective reinforcement learning (MORL) seeks to train agents capable of balancing conflicting objectives.
Mitigating social bias in Large Language Models (LLMs) presents a distinct alignment challenge: unlike verifiable tasks, bias lacks a single ground truth, creating a high-variance, subjective reward landscape. Previous preference-based fine-tuning methods have major trade-offs: Direct Preference Optimization (DPO) is limited by the lack of exploration inherent in offline training, while Proximal Policy Optimization (PPO) can lead to training instability due to potentially unreliable critic estimates.
arXiv:2607. 20515v1 Announce Type: new Abstract: Reinforcement Learning from Human Feedback (RLHF) is critical for aligning Large Language Models (LLMs) with human preferences.
arXiv:2510. 05342v2 Announce Type: replace-cross Abstract: Direct Preference Optimization (DPO) has emerged as a simple and effective method for aligning large language models.
arXiv:2606. 01382v1 Announce Type: cross Abstract: Preference alignment is central to improving large language models, but standard reward-based formulations can be restrictive when human preferences are cyclic, non-transitive, or otherwise not representable by a scalar reward.
arXiv:2606. 07678v1 Announce Type: cross Abstract: Safety alignment for large language models relies on preference data, but current pipelines often train on large, redundant datasets.