RefEdit, an instruction-based editing model trained on synthetic data, outperforms baselines in complex scene editing and referring expression tasks.
Despite recent advances in inversion and instruction-based image editing,
existing approaches primarily excel at editing single, prominent objects but
significantly struggle when applied to complex scenes containing multiple
entities. To quantify this gap, we first introduce RefEdit-Bench, a rigorous
real-world benchmark rooted in RefCOCO, where even baselines trained on
millions of samples perform poorly. To overcome this limitation, we introduce
RefEdit — an instruction-based editing model trained on our scalable synthetic
data generation pipeline. Our RefEdit, trained on only 20,000 editing triplets,
outperforms the Flux/SD3 model-based baselines trained on millions of data.
Extensive evaluations across various benchmarks demonstrate that our model not
only excels in referring expression tasks but also enhances performance on
traditional benchmarks, achieving state-of-the-art results comparable to
closed-source methods. We release data \& checkpoint for reproducibility.