Search papers, labs, and topics across Lattice.
The paper introduces VQ-VA World, a data-centric framework with an agentic pipeline, to generate high-quality image-text samples for training Visual Question-Visual Answering (VQ-VA) models. They crawled ~1.8M interleaved image-text samples and created IntelligentBench, a human-curated benchmark, to evaluate VQ-VA models on world knowledge, design knowledge, and reasoning. Training LightFusion with VQ-VA World data significantly improves performance on IntelligentBench, closing the gap with proprietary systems like NanoBanana and GPT-Image.
Open-source VQ-VA models just got a massive boost: a new dataset and benchmark close the gap with proprietary systems on visual question-visual answering.
This paper studies Visual Question-Visual Answering (VQ-VA): generating an image, rather than text, in response to a visual question -- an ability that has recently emerged in proprietary systems such as NanoBanana and GPT-Image. To also bring this capability to open-source models, we introduce VQ-VA World, a data-centric framework built around an agentic pipeline for large-scale, targeted data construction. Leveraging web-scale deployment, this pipeline crawls a massive amount of ~1.8M high-quality, interleaved image-text samples for model training. For evaluation, we further release IntelligentBench, a human-curated benchmark that systematically assesses VQ-VA along the aspects of world knowledge, design knowledge, and reasoning. Training with VQ-VA World data yields strong empirical gains: it helps LightFusion attain 53.06 on IntelligentBench, substantially surpassing the best prior open-source baselines (i.e., 7.78 from vanilla LightFusion; 1.94 from UniWorld-V1), and significantly narrowing the gap toward leading proprietary systems (e.g., 81.67 from NanoBanana; 82.64 from GPT-Image). By releasing the full suite of model weights, datasets, and pipelines, we hope to stimulate future research on VQ-VA.
