Same Layout, Different Count: What Number Words Do Past Ten in Text-to-Image Models
cs.CV, cs.LG
Submitted: 2026-11-04
Updated: 2026-11-04
Terminology
Sources
- Qwen2.5-VL Technical Report
- eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers
- CountSteer: Steering Attention for Object Counting in Diffusion Models
- Temporal Concept Dynamics in Diffusion Models via Prompt-Conditioned Interventions
- Text-to-Image Diffusion Models Cannot Count, and Prompt Refinement Cannot Help
- Classifier-Free Diffusion Guidance
- GenEval 2: Addressing Benchmark Drift in Text-to-Image Evaluation
- ComposeAnything: Composite Object Priors for Text-to-Image Generation
- CountCluster: Training-Free Object Quantity Guidance with Cross-Attention Map Clustering for Text-to-Image Generation
- Faster Diffusion via Temporal Attention Decomposition
- DifFRACT: Diffusion Feature Reconstruction and Attribution for Circuit Tracing
- CountLoop: Training-Free High-Instance Image Generation via Iterative Agent Guidance
- Learning to Generate Multiple Objects from Dense and Occluded Layouts
- Flow-DPPO: Divergence Proximal Policy Optimization for Flow Matching Models
- ATHENA: Adaptive Test-Time Steering for Improving Count Fidelity in Diffusion Models
- Contrastive Prompts Improve Disentanglement in Text-to-Image Diffusion Models
- Qwen-Image Technical Report
- Qwen3 Technical Report
- Demystifying Numerosity in Diffusion Models -- Limitations and Remedies
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