RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Oh, Yeongtak, Chung, Dohyun, Shin, Juhyeon, Park, Sangha, Barthelemy, Johan, Mok, Jisoo, Yoon, Sungroh
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866909833571074048
author Oh, Yeongtak
Chung, Dohyun
Shin, Juhyeon
Park, Sangha
Barthelemy, Johan
Mok, Jisoo
Yoon, Sungroh
author_facet Oh, Yeongtak
Chung, Dohyun
Shin, Juhyeon
Park, Sangha
Barthelemy, Johan
Mok, Jisoo
Yoon, Sungroh
contents Recent multi-modal large language models (MLLMs) often struggle to generate personalized image captions, even when trained on high-quality captions. In this work, we observe that such limitations persist in existing post-training-based MLLM personalization methods. Specifically, despite being post-tuned with large-scale caption data through supervised fine-tuning (SFT), these models frequently fail to produce faithful descriptions in real-world scenarios, such as multi-concept image captioning. However, acquiring large-scale, high-quality captions for such complex settings is both costly and difficult. To address the data-centric nature of SFT, we propose a reinforcement learning (RL)-based post-training framework. To the best of our knowledge, this is the first RL-based approach to post-train MLLMs for personalized image captioning. Our method significantly enhances both visual recognition and personalized generation capabilities of MLLMs, and consistently outperforms existing SFT-based baselines, especially in the challenging multi-concept image captioning task. Project page: https://github.com/oyt9306/RePIC
format Preprint
id arxiv_https___arxiv_org_abs_2506_18369
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models
Oh, Yeongtak
Chung, Dohyun
Shin, Juhyeon
Park, Sangha
Barthelemy, Johan
Mok, Jisoo
Yoon, Sungroh
Computer Vision and Pattern Recognition
Recent multi-modal large language models (MLLMs) often struggle to generate personalized image captions, even when trained on high-quality captions. In this work, we observe that such limitations persist in existing post-training-based MLLM personalization methods. Specifically, despite being post-tuned with large-scale caption data through supervised fine-tuning (SFT), these models frequently fail to produce faithful descriptions in real-world scenarios, such as multi-concept image captioning. However, acquiring large-scale, high-quality captions for such complex settings is both costly and difficult. To address the data-centric nature of SFT, we propose a reinforcement learning (RL)-based post-training framework. To the best of our knowledge, this is the first RL-based approach to post-train MLLMs for personalized image captioning. Our method significantly enhances both visual recognition and personalized generation capabilities of MLLMs, and consistently outperforms existing SFT-based baselines, especially in the challenging multi-concept image captioning task. Project page: https://github.com/oyt9306/RePIC
title RePIC: Reinforced Post-Training for Personalizing Multi-Modal Language Models
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.18369