paper-with-me

Papers

Human Preference Score v2: A Solid Benchmark for Evaluating Human Preferences of Text-to-Image Synthesis

2023-06-15 · Xiaoshi Wu, Yiming Hao, Keqiang Sun, Yixiong Chen, Feng Zhu, Rui Zhao, Hongsheng Li

Recent text-to-image generative models can generate high-fidelity images from text inputs, but the quality of these generated images cannot be accurately evaluated by existing evaluation metrics. To address this issue, we introduce Human Preference Dataset v2 (HPD v2), a large-scale dataset that captures human preferences on images from a wide range of sources. HPD v2 comprises 798,090 human preference choices on 433,760 pairs of images, making it the largest dataset of its kind. The text prompts and images are deliberately collected to eliminate potential bias, which is a common issue in previous datasets. By fine-tuning CLIP on HPD v2, we obtain Human Preference Score v2 (HPS v2), a scoring model that can more accurately predict human preferences on generated images. Our experiments demonstrate that HPS v2 generalizes better than previous metrics across various image distributions and is responsive to algorithmic improvements of text-to-image generative models, making it a preferable evaluation metric for these models. We also investigate the design of the evaluation prompts for text-to-image generative models, to make the evaluation stable, fair and easy-to-use. Finally, we establish a benchmark for text-to-image generative models using HPS v2, which includes a set of recent text-to-image models from the academic, community and industry. The code and dataset is available at https://github.com/tgxs002/HPSv2 .

📄 PDF Abstract BibTeX arXiv:2306.09341

Code (1)

tgxs002/hpsv2 공식 구현 pytorch

Tasks

Image GenerationPreference Mapping

Methods 이 논문이 사용한 방법론

CLIP Contrastive Language-Image Pre-training (CLIP), consisting of a simplified version of ConVIRT trained from scratch, is an efficient method of image representation learning…

Similar Papers 제목 키워드 기반

Many Voices, One Reward: Multi-Role Rubric Generation for LLM Judging and Reward Modeling

2026-07-02 · Dazhi Fu, Jiuding Yang, Yiwen Guo, Jicong Fan arxiv

Reliable reward and preference signals are critical for evaluating and optimizing large language models on open-ended tasks. Rubric-based judges offer a transparent way to decompose such judgments into explicit evaluatio…

Reinforcement Learning

Human Preference-Aligned Concept Customization Benchmark via Decomposed Evaluation

2025-09-03 · Reina Ishikawa, Ryo Fujii, Hideo Saito, Ryo Hachiuma arxiv

Evaluating concept customization is challenging, as it requires a comprehensive assessment of fidelity to generative prompts and concept images. Moreover, evaluating multiple concepts is considerably more difficult than …

Establishing Reliability Metrics for Reward Models in Large Language Models

2025-04-21 · Yizhou Chen, Yawen Liu, Xuesi Wang, Qingtao Yu 외

The reward model (RM) that represents human preferences plays a crucial role in optimizing the outputs of large language models (LLMs), e.g., through reinforcement learning from human feedback (RLHF) or rejection samplin…

Benchmarking

Evaluating Text-to-Image Synthesis with a Conditional Fréchet Distance

2025-03-27 · Jaywon Koo, Jefferson Hernandez, Moayed Haji-Ali, Ziyan Yang 외

Evaluating text-to-image synthesis is challenging due to misalignment between established metrics and human preferences. We propose cFreD, a metric based on the notion of Conditional Fr\'echet Distance that explicitly ac…

BenchmarkingImage Generation

HP-Edit: A Human-Preference Post-Training Framework for Image Editing

2026-04-21 · Fan Li, Chonghuinan Wang, Lina Lei, Yuping Qiu 외 arxiv

Common image editing tasks typically adopt powerful generative diffusion models as the leading paradigm for real-world content editing. Meanwhile, although reinforcement learning (RL) methods such as Diffusion-DPO and Fl…

Reinforcement LearningImage Editing