DreamBench++: A Human-Aligned Benchmark for Personalized Image Generation
This addresses the need for efficient and accurate evaluation in personalized image generation, which is incremental as it builds on existing automated methods to improve alignment with human judgments.
The authors tackled the problem of evaluating personalized image generation models by introducing DreamBench++, a human-aligned benchmark that uses advanced multimodal GPT models to automate evaluations, which they demonstrated by benchmarking 7 modern generative models to achieve more human-aligned results.
Personalized image generation holds great promise in assisting humans in everyday work and life due to its impressive ability to creatively generate personalized content across various contexts. However, current evaluations either are automated but misalign with humans or require human evaluations that are time-consuming and expensive. In this work, we present DreamBench++, a human-aligned benchmark that advanced multimodal GPT models automate. Specifically, we systematically design the prompts to let GPT be both human-aligned and self-aligned, empowered with task reinforcement. Further, we construct a comprehensive dataset comprising diverse images and prompts. By benchmarking 7 modern generative models, we demonstrate that DreamBench++ results in significantly more human-aligned evaluation, helping boost the community with innovative findings.