GenScale: A Benchmark for Relative Object Scale in Image Generation and Editing
cs.CV
Submitted: 2026-09-01
Updated: 2026-11-06
Code: https://github.com/Tongyi-MAI/Z-Image
Project page: https://lingxiao-li.github.io/genscale.github.io/GPT
Terminology
Sources
- Commonsense-T2I Challenge: Can Text-to-Image Generation Models Understand Commonsense?
- Generate Any Scene: Scene Graph Driven Data Synthesis for Visual Generation Training
- GenAI-Bench: Evaluating and Improving Compositional Text-to-Visual Generation
- Evaluating Text-to-Visual Generation with Image-to-Text Generation
- PhyBench: A Physical Commonsense Benchmark for Evaluating Text-to-Image Models
- WISE: A World Knowledge-Informed Semantic Evaluation for Text-to-Image Generation
- Hierarchical Text-Conditional Image Generation with CLIP Latents
- SAM 2: Segment Anything in Images and Videos
- Insert Anything: Image Insertion via In-Context Editing in DiT
- Everything in Its Place: Benchmarking Spatial Intelligence of Text-to-Image Models
- Depth Anything V2
- IP-Adapter: Text Compatible Image Prompt Adapter for Text-to-Image Diffusion Models
- Scaling Autoregressive Models for Content-Rich Text-to-Image Generation
- A-Bench: Are LMMs Masters at Evaluating AI-generated Images?
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs