Selecting Fine-Tuning Examples by Quizzing VLMs

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ji, Tenghao, Adar, Eytan
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866912710563725312
author Ji, Tenghao
Adar, Eytan
author_facet Ji, Tenghao
Adar, Eytan
contents A challenge in fine-tuning text-to-image diffusion models for specific topics is to select good examples. Fine-tuning from image sets of varying quality, such as Wikipedia Commons, will often produce poor output. However, training images that \textit{do} exemplify the target concept (e.g., a \textit{female Mountain Bluebird}) help ensure that the generated images are similarly representative (e.g., have the prototypical blue-wings and gray chest). In this work, we propose QZLoRA, a framework to select images for low-rank adaptation (LoRA). The approach leverages QuizRank, a method to automatically rank images by treating them as an `educational intervention' and `quizzing' a VLM. We demonstrate that QZLoRA can produce better aligned, photorealistic images with fewer samples. We also show that these fine-tuned models can produce stylized that are similarly representative (i.e., illustrations). Our results highlight the promise of combining automated visual reasoning with parameter-efficient fine-tuning for topic-adaptive generative modeling.
format Preprint
id arxiv_https___arxiv_org_abs_2511_12002
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Selecting Fine-Tuning Examples by Quizzing VLMs
Ji, Tenghao
Adar, Eytan
Machine Learning
Computer Vision and Pattern Recognition
A challenge in fine-tuning text-to-image diffusion models for specific topics is to select good examples. Fine-tuning from image sets of varying quality, such as Wikipedia Commons, will often produce poor output. However, training images that \textit{do} exemplify the target concept (e.g., a \textit{female Mountain Bluebird}) help ensure that the generated images are similarly representative (e.g., have the prototypical blue-wings and gray chest). In this work, we propose QZLoRA, a framework to select images for low-rank adaptation (LoRA). The approach leverages QuizRank, a method to automatically rank images by treating them as an `educational intervention' and `quizzing' a VLM. We demonstrate that QZLoRA can produce better aligned, photorealistic images with fewer samples. We also show that these fine-tuned models can produce stylized that are similarly representative (i.e., illustrations). Our results highlight the promise of combining automated visual reasoning with parameter-efficient fine-tuning for topic-adaptive generative modeling.
title Selecting Fine-Tuning Examples by Quizzing VLMs
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2511.12002