ImgEdit supplies 1.2 million curated edit pairs and a three-part benchmark that let a VLM-based model outperform prior open-source editors on adherence, quality, and detail preservation.
Improving image captioning with better use of captions
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
citation-role summary
baseline 1
method 1
citation-polarity summary
fields
cs.CV 4representative citing papers
Iterative self-improving codebooks enhance safety in autoregressive multimodal models by self-identifying unsafe generations and updating the codebook to eliminate harmful visual token mappings without external feedback.
UniRect-CoT is a training-free rectification chain-of-thought framework that treats diffusion denoising as visual reasoning and uses the model's inherent understanding to align and correct intermediate generation results.
UniWorld-V1 shows that semantic features from large multimodal models enable unified visual understanding and generation, achieving strong results on perception and manipulation tasks with only 2.7 million training samples.
citing papers explorer
-
ImgEdit: A Unified Image Editing Dataset and Benchmark
ImgEdit supplies 1.2 million curated edit pairs and a three-part benchmark that let a VLM-based model outperform prior open-source editors on adherence, quality, and detail preservation.
-
Safe Autoregressive Image Generation with Iterative Self-Improving Codebooks
Iterative self-improving codebooks enhance safety in autoregressive multimodal models by self-identifying unsafe generations and updating the codebook to eliminate harmful visual token mappings without external feedback.
-
Free Lunch for Unified Multimodal Models: Enhancing Generation via Reflective Rectification with Inherent Understanding
UniRect-CoT is a training-free rectification chain-of-thought framework that treats diffusion denoising as visual reasoning and uses the model's inherent understanding to align and correct intermediate generation results.
-
UniWorld-V1: High-Resolution Semantic Encoders for Unified Visual Understanding and Generation
UniWorld-V1 shows that semantic features from large multimodal models enable unified visual understanding and generation, achieving strong results on perception and manipulation tasks with only 2.7 million training samples.