paper-with-me

Papers

Diverse Preference Optimization

2025-01-30 · Jack Lanchantin, Angelica Chen, Shehzaad Dhuliawala, Ping Yu, Jason Weston, Sainbayar Sukhbaatar, Ilia Kulikov

Post-training of language models, either through reinforcement learning, preference optimization or supervised finetuning, tends to sharpen the output probability distribution and reduce the diversity of generated responses. This is particularly a problem for creative generative tasks where varied responses are desired. In this work we introduce Diverse Preference Optimization (DivPO), an optimization method which learns to generate much more diverse responses than standard pipelines, while maintaining the quality of the generations. In DivPO, preference pairs are selected by first considering a pool of responses, and a measure of diversity among them, and selecting chosen examples as being more rare but high quality, while rejected examples are more common, but low quality. DivPO results in generating 45.6% more diverse persona attributes, and an 74.6% increase in story diversity, while maintaining similar win rates as standard baselines.

📄 PDF Abstract BibTeX arXiv:2501.18101

Code (1)

mj-storytelling/DiversityTuning pytorch

Tasks

Diversity

Similar Papers 제목 키워드 기반

BAPO: Base-Anchored Preference Optimization for Overcoming Forgetting in Large Language Models Personalization

2024-06-30 · Gihun Lee, Minchan Jeong, Yujin Kim, Hojung Jung 외

While learning to align Large Language Models (LLMs) with human preferences has shown remarkable success, aligning these models to meet the diverse user preferences presents further challenges in preserving previous know…

Continual LearningGeneral KnowledgeKnowledge DistillationPersonalized and Emotional Conversation+1

Relative Preference Optimization: Enhancing LLM Alignment through Contrasting Responses across Identical and Diverse Prompts

2024-02-12 · Yueqin Yin, Zhendong Wang, Yi Gu, Hai Huang 외

In the field of large language models (LLMs), aligning models with the diverse preferences of users is a critical challenge. Direct Preference Optimization (DPO) has played a key role in this area. It works by using pair…

FairPO: Robust Preference Optimization for Fair Multi-Label Learning

2025-05-05 · Soumen Kumar Mondal, Akshit Varmora, Prateek Chanda, Ganesh Ramakrishnan

We propose FairPO, a novel framework designed to promote fairness in multi-label classification by directly optimizing preference signals with a group robustness perspective. In our framework, the set of labels is partit…

FairnessMulti-Label ClassificationMUlTI-LABEL-ClASSIFICATIONMulti-Label Learning

Annotation-Efficient Preference Optimization for Language Model Alignment

2024-05-22 · Yuu Jinnai, Ukyo Honda

Preference optimization is a standard approach to fine-tuning large language models to align with human preferences. The quality, diversity, and quantity of the preference dataset are critical to the effectiveness of pre…

DiversityLanguage ModelingLanguage Modellingmodel

Direct Preference Optimization With Unobserved Preference Heterogeneity

2024-05-23 · Keertana Chidambaram, Karthik Vinay Seetharaman, Vasilis Syrgkanis

RLHF has emerged as a pivotal step in aligning language models with human objectives and values. It typically involves learning a reward model from human preference data and then using reinforcement learning to update th…

Ensemble Learningreinforcement-learningReinforcement Learning