paper-with-me

홈 › Papers

Personalized Adaptation via In-Context Preference Learning

2024-10-17 · Allison Lau, Younwoo Choi, Vahid Balazadeh, Keertana Chidambaram, Vasilis Syrgkanis, Rahul G. Krishnan

Reinforcement Learning from Human Feedback (RLHF) is widely used to align Language Models (LMs) with human preferences. However, existing approaches often neglect individual user preferences, leading to suboptimal personalization. We present the Preference Pretrained Transformer (PPT), a novel approach for adaptive personalization using online user feedback. PPT leverages the in-context learning capabilities of transformers to dynamically adapt to individual preferences. Our approach consists of two phases: (1) an offline phase where we train a single policy model using a history-dependent loss function, and (2) an online phase where the model adapts to user preferences through in-context learning. We demonstrate PPT's effectiveness in a contextual bandit setting, showing that it achieves personalized adaptation superior to existing methods while significantly reducing the computational costs. Our results suggest the potential of in-context learning for scalable and efficient personalization in large language models.

📄 PDF Abstract BibTeX arXiv:2410.14001

Code (0)

등록된 구현이 없습니다.

Tasks

In-Context Learning

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Residual Connection 설명 없음
Position-Wise Feed-Forward Layer 설명 없음
Adam 설명 없음
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…

Similar Papers 제목 키워드 기반

From 1,000,000 Users to Every User: Scaling Up Personalized Preference for User-level Alignment

2025-03-19 · Jia-Nan Li, Jian Guan, Songhao Wu, Wei Wu 외

Large language models (LLMs) have traditionally been aligned through one-size-fits-all approaches that assume uniform human preferences, fundamentally overlooking the diversity in user values and needs. This paper introd…

Diversity

Learning Color Grading, No Photo Sharing: Federated Aesthetic Preference Learning for Personalized Image Enhancement

2026-07-30 · Chuanzhi Xu, Ziyuan Tao, Jean Julien KNell, Yanrong Chen 외 arxiv

Personalized image enhancement should reflect individual aesthetic taste, yet learning such preferences commonly depends on private photos and ratings that are unsuitable for centralized collection. The task must infer p…

Image Enhancement

One Adapts to Any: Meta Reward Modeling for Personalized LLM Alignment

2026-01-26 · Hongru Cai, Yongqi Li, Tiezheng Yu, Fengbin Zhu 외 arxiv

Alignment of Large Language Models (LLMs) aims to align outputs with human preferences, and personalized alignment further adapts models to individual users. This relies on personalized reward models that capture user-sp…

FSPO: Few-Shot Preference Optimization of Synthetic Preference Data in LLMs Elicits Effective Personalization to Real Users

2025-02-26 · Anikait Singh, Sheryl Hsu, Kyle Hsu, Eric Mitchell 외

Effective personalization of LLMs is critical for a broad range of user-interfacing applications such as virtual assistants and content curation. Inspired by the strong in-context learning capabilities of LLMs, we propos…

In-Context LearningMeta-LearningOpen-Ended Question AnsweringQuestion Answering

The Reward Model Selection Crisis in Personalized Alignment

2025-12-28 · Fady Rezk, Yuangang Pan, Chuan-Sheng Foo, Xun Xu 외 arxiv

Personalized alignment from preference data has focused primarily on improving personal reward model (RM) accuracy, with the implicit assumption that better preference ranking translates to better personalized behavior. …