EPISODE · Mar 28, 2026 · 22 MIN
MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data
from Daily Paper Cast · host Jingwen Liang, Gengyu Wang
🤗 Upvotes: 26 | cs.CV Authors: Zhekai Chen, Yuqing Wang, Manyuan Zhang, Xihui Liu Title: MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data Arxiv: http://arxiv.org/abs/2603.25319v1 Abstract: Generating images conditioned on multiple visual references is critical for real-world applications such as multi-subject composition, narrative illustration, and novel view synthesis, yet current models suffer from severe performance degradation as the number of input references grows. We identify the root cause as a fundamental data bottleneck: existing datasets are dominated by single- or few-reference pairs and lack the structured, long-context supervision needed to learn dense inter-reference dependencies. To address this, we introduce MacroData, a large-scale dataset of 400K samples, each containing up to 10 reference images, systematically organized across four complementary dimensions -- Customization, Illustration, Spatial reasoning, and Temporal dynamics -- to provide comprehensive coverage of the multi-reference generation space. Recognizing the concurrent absence of standardized evaluation protocols, we further propose MacroBench, a benchmark of 4,000 samples that assesses generative coherence across graded task dimensions and input scales. Extensive experiments show that fine-tuning on MacroData yields substantial improvements in multi-reference generation, and ablation studies further reveal synergistic benefits of cross-task co-training and effective strategies for handling long-context complexity. The dataset and benchmark will be publicly released.
What this episode covers
🤗 Upvotes: 26 | cs.CV Authors: Zhekai Chen, Yuqing Wang, Manyuan Zhang, Xihui Liu Title: MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data Arxiv: http://arxiv.org/abs/2603.25319v1 Abstract: Generating images conditioned on multiple visual references is critical for real-world applications such as multi-subject composition, narrative illustration, and novel view synthesis, yet current models suffer from severe performance degradation as the number of input references grows. We identify the root cause as a fundamental data bottleneck: existing datasets are dominated by single- or few-reference pairs and lack the structured, long-context supervision needed to learn dense inter-reference dependencies. To address this, we introduce MacroData, a large-scale dataset of 400K samples, each containing up to 10 reference images, systematically organized across four complementary dimensions -- Customization, Illustration, Spatial reasoning, and Temporal dynamics -- to provide comprehensive coverage of the multi-reference generation space. Recognizing the concurrent absence of standardized evaluation protocols, we further propose MacroBench, a benchmark of 4,000 samples that assesses generative coherence across graded task dimensions and input scales. Extensive experiments show that fine-tuning on MacroData yields substantial improvements in multi-reference generation, and ablation studies further reveal synergistic benefits of cross-task co-training and effective strategies for handling long-context complexity. The dataset and benchmark will be publicly released.
NOW PLAYING
MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data
No transcript for this episode yet
Similar Episodes
No similar episodes found.