# Don't Start Over: A Cost-Effective Framework for Migrating Personalized Prompts Between LLMs

arXiv: https://arxiv.org/abs/2601.12034 · HTML: https://mlsanity.com/paper/2601.12034/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 32 references that are in the map.
- First posted on arXiv 17 Jan 2026.

## Builds on (most cited first, 20 of 20)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Beyond neural scaling laws: beating power law scaling via data pruning](https://mlsanity.com/paper/2206.14486/) (2022)
- [Selection via Proxy: Efficient Data Selection for Deep Learning](https://mlsanity.com/paper/1906.11829/) (2019)
- [Personalized Soups: Personalized Large Language Model Alignment via Post-hoc Parameter Merging](https://mlsanity.com/paper/2310.11564/) (2023)
- [SPoT: Better Frozen Model Adaptation through Soft Prompt Transfer](https://mlsanity.com/paper/2110.07904/) (2021)
- [Is ChatGPT a Good Recommender? A Preliminary Study](https://mlsanity.com/paper/2304.10149/) (2023)
- [Large Language Models for Education: A Survey and Outlook](https://mlsanity.com/paper/2403.18105/) (2024)
- [Coverage-centric Coreset Selection for High Pruning Rates](https://mlsanity.com/paper/2210.15809/) (2022)
- [Do LLMs Understand User Preferences? Evaluating LLMs On User Rating Prediction](https://mlsanity.com/paper/2305.06474/) (2023)
- [CoLLM: Integrating Collaborative Embeddings into Large Language Models for Recommendation](https://mlsanity.com/paper/2310.19488/) (2023)
- [Personalized Prompt Learning for Explainable Recommendation](https://mlsanity.com/paper/2202.07371/) (2022)
- [Adaptive Second Order Coresets for Data-efficient Machine Learning](https://mlsanity.com/paper/2207.13887/) (2022)
- [Large Language Models are Learnable Planners for Long-Term Recommendation](https://mlsanity.com/paper/2403.00843/) (2024)
- [SPRec: Self-Play to Debias LLM-based Recommendation](https://mlsanity.com/paper/2412.09243/) (2024)
- [Process-Supervised LLM Recommenders via Flow-guided Tuning](https://mlsanity.com/paper/2503.07377/) (2025)
- [Fine-grained List-wise Alignment for Generative Medication Recommendation](https://mlsanity.com/paper/2505.20218/) (2025)

## Built on by (most cited first, 3 of 3)

- [Breaking User-Centric Agency: A Tri-Party Framework for Agent-Based Recommendation](https://mlsanity.com/paper/2603.10673/) (2026)
- [Towards Root Memories: Benchmarking and Enhancing Implicit Logical Memory Retrieval for Personalized LLMs](https://mlsanity.com/paper/2606.23283/) (2026)
- [Position-Aware Drafting for Inference Acceleration in LLM-Based Generative List-Wise Recommendation](https://mlsanity.com/paper/2604.27747/) (2026)

## Abstract (arXiv)

Personalization in Large Language Models (LLMs) often relies on user-specific soft prompts. However, these prompts become obsolete when the foundation model is upgraded, necessitating costly, full-scale retraining. To overcome this limitation, we propose the Prompt-level User Migration Adapter (PUMA), a lightweight framework to efficiently migrate personalized prompts across incompatible models. PUMA utilizes a parameter-efficient adapter to bridge the semantic gap, combined with a group-based user selection strategy to significantly reduce training costs. Experiments on three large-scale datasets show our method matches or even surpasses the performance of retraining from scratch, reducing computational cost by up to 98%. The framework demonstrates strong generalization across diverse model architectures and robustness in advanced scenarios like chained and aggregated migrations, offering a practical path for the sustainable evolution of personalized AI by decoupling user assets from the underlying models.