PROPER: A Progressive Learning Framework for Personalized Large Language Models with Group-Level Adaptation
Fuente:
arXiv
Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866913715307151360 |
|---|---|
| author | Zhang, Linhai Wu, Jialong Zhou, Deyu He, Yulan |
| author_facet | Zhang, Linhai Wu, Jialong Zhou, Deyu He, Yulan |
| contents | Personalized large language models (LLMs) aim to tailor their outputs to user preferences. Recent advances in parameter-efficient fine-tuning (PEFT) methods have highlighted the effectiveness of adapting population-level LLMs to personalized LLMs by fine-tuning user-specific parameters with user history. However, user data is typically sparse, making it challenging to adapt LLMs to specific user patterns. To address this challenge, we propose PROgressive PERsonalization (PROPER), a novel progressive learning framework inspired by meso-level theory in social science. PROPER bridges population-level and user-level models by grouping users based on preferences and adapting LLMs in stages. It combines a Mixture-of-Experts (MoE) structure with Low Ranked Adaptation (LoRA), using a user-aware router to assign users to appropriate groups automatically. Additionally, a LoRA-aware router is proposed to facilitate the integration of individual user LoRAs with group-level LoRAs. Experimental results show that PROPER significantly outperforms SOTA models across multiple tasks, demonstrating the effectiveness of our approach. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_01303 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | PROPER: A Progressive Learning Framework for Personalized Large Language Models with Group-Level Adaptation Zhang, Linhai Wu, Jialong Zhou, Deyu He, Yulan Computation and Language Personalized large language models (LLMs) aim to tailor their outputs to user preferences. Recent advances in parameter-efficient fine-tuning (PEFT) methods have highlighted the effectiveness of adapting population-level LLMs to personalized LLMs by fine-tuning user-specific parameters with user history. However, user data is typically sparse, making it challenging to adapt LLMs to specific user patterns. To address this challenge, we propose PROgressive PERsonalization (PROPER), a novel progressive learning framework inspired by meso-level theory in social science. PROPER bridges population-level and user-level models by grouping users based on preferences and adapting LLMs in stages. It combines a Mixture-of-Experts (MoE) structure with Low Ranked Adaptation (LoRA), using a user-aware router to assign users to appropriate groups automatically. Additionally, a LoRA-aware router is proposed to facilitate the integration of individual user LoRAs with group-level LoRAs. Experimental results show that PROPER significantly outperforms SOTA models across multiple tasks, demonstrating the effectiveness of our approach. |
| title | PROPER: A Progressive Learning Framework for Personalized Large Language Models with Group-Level Adaptation |
| topic | Computation and Language |
| url | https://arxiv.org/abs/2503.01303 |