CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2025
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866911423319244800 |
|---|---|
| author | Lv, Hang Liang, Sheng Wang, Hao Gu, Hongchao Wu, Yaxiong Guo, Wei Lian, Defu Liu, Yong Chen, Enhong |
| author_facet | Lv, Hang Liang, Sheng Wang, Hao Gu, Hongchao Wu, Yaxiong Guo, Wei Lian, Defu Liu, Yong Chen, Enhong |
| contents | Personalization has become crucial for adapting models to the diverse and evolving needs of users across cultural, temporal, and contextual dimensions. While existing methods often rely on centralized fine-tuning or static preference alignment within a single model, they struggle to achieve both real-time and high-quality personalization under the resource and privacy constraints of personal devices. To address this challenge, we propose CoSteer, a collaborative framework that enables tuning-free, real-time personalization via decoding-time adaptation. By leveraging logit differences between context-aware and context-agnostic local small models, CoSteer steers cloud-based large models, ensuring effective personalization while preserving the large model's capabilities. Personalization is handled locally, with only final tokens sent to the cloud, maintaining both user context and system efficiency. Through extensive experiments across a wide range of tasks, we demonstrate that CoSteer generates high-quality personalized content, ensuring both effectiveness and computational efficiency. Our results highlight its robustness across models and environments, confirming its practical applicability in real-world scenarios. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2507_04756 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering Lv, Hang Liang, Sheng Wang, Hao Gu, Hongchao Wu, Yaxiong Guo, Wei Lian, Defu Liu, Yong Chen, Enhong Computation and Language Artificial Intelligence Personalization has become crucial for adapting models to the diverse and evolving needs of users across cultural, temporal, and contextual dimensions. While existing methods often rely on centralized fine-tuning or static preference alignment within a single model, they struggle to achieve both real-time and high-quality personalization under the resource and privacy constraints of personal devices. To address this challenge, we propose CoSteer, a collaborative framework that enables tuning-free, real-time personalization via decoding-time adaptation. By leveraging logit differences between context-aware and context-agnostic local small models, CoSteer steers cloud-based large models, ensuring effective personalization while preserving the large model's capabilities. Personalization is handled locally, with only final tokens sent to the cloud, maintaining both user context and system efficiency. Through extensive experiments across a wide range of tasks, we demonstrate that CoSteer generates high-quality personalized content, ensuring both effectiveness and computational efficiency. Our results highlight its robustness across models and environments, confirming its practical applicability in real-world scenarios. |
| title | CoSteer: Collaborative Decoding-Time Personalization via Local Delta Steering |
| topic | Computation and Language Artificial Intelligence |
| url | https://arxiv.org/abs/2507.04756 |