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.
By Hyung Gyu Rho
arXiv:2504.20106v4 Announce Type: replace
Abstract: Ensuring that large language models (LLMs) are both helpful and harmless is a critical challenge, as overly strict constraints can lead to excessiv...
By Ren-Wei Liang, Chin-Ting Hsu, Chan-Hung Yu, Saransh Agrawal, Shih-Cheng Huang, Chieh-Yen Lin, Shang-Tse Chen, Kuan-Hao Huang, Shao-Hua Sun
arXiv:2608. 09507v1 Announce Type: cross Abstract: Natural language user preferences provide an interpretable interface for LLM personalization.
By Yuting Liu, Wei Wu, Jianzhe Zhao, Guibing Guo
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.
By Saket Reddy, Ke Yang, ChengXiang Zhai
The paper introduces GAW-PO, a gradient‑aligned token reweighting technique for Direct Preference Optimization (DPO). It assigns weaker penalties to rejected tokens whose gradients align with preferred behavior, while maintaining stronger penalties for conflicting tokens. Across 11 benchmarks in mathematics, reasoning, coding, and question answering, GAW‑PO outperforms standard DPO and other baselines, and remains robust when the DPO regularization parameter is reduced.
By Andreea Dutulescu, Stefan Ruseti, Mihai Masala, Traian Rebedea, Mihai Dascalu
The paper introduces Personalized Group Relative Policy Optimization (P‑GRPO), a new alignment framework for large language models that separates advantage estimation from immediate batch statistics. By normalizing advantages using preference‑group‑specific reward histories instead of the concurrent generation group, P‑GRPO maintains contrastive signals for distinct user preferences. Experiments across various tasks show that P‑GRPO converges faster and yields higher rewards than standard GRPO, improving alignment with heterogeneous human preferences while preserving general capabilities.
By Jialu Wang, Heinrich Peters, Asad A. Butt, Navid Hashemi, Alireza Hashemi, Pouya M. Ghari, Joseph Hoover, James Rae, Morteza Dehghani