Skip to content
Mobrief
Mobrief
Back to archive

Research · arXiv cs.CV

When Numbers Speak: Aligning Textual Numerals and Visual Instances in Text-to-Video Diffusion Models

Text-to-video diffusion models have enabled open-ended video synthesis, but often struggle with generating the correct number of objects specified in a prompt.

Apr 09, 2026 17:59 UTC · Paper: ~15 min · Research
Research

Check methodology

Read original
  • arXiv cs.CV introduces NUMINA , a training-free identify-then-guide framework for improved numerical alignment.
  • NUMINA identifies prompt-layout inconsistencies by selecting discriminative self- and cross-attention heads to derive a countable latent layout.
  • It then refines this layout conservatively and modulates cross-attention to guide regeneration.

Context

arXiv cs.CV introduces NUMINA , a training-free identify-then-guide framework for improved numerical alignment. NUMINA identifies prompt-layout inconsistencies by selecting discriminative self- and cross-attention heads to derive a countable latent layout. It then refines this layout conservatively and modulates cross-attention to guide regeneration. On the introduced CountBench, NUMINA improves counting accuracy by up to 7.4% on Wan2.1-1.3B, and by 4.9% and 5.5% on 5B and 14B models, respectively. Furthermore, CLIP alignment is improved while maintaining temporal consistency. These results demonstrate that structural guidance complements seed search and prompt enhancement, offering a practical path toward count-accurate text-to-video diffusion. The code is available at https://github.com/H-EmbodVis/NUMINA.

For builders

arXiv cs.CV introduces NUMINA , a training-free identify-then-guide framework for improved numerical alignment.

arXiv cs.CV introduces NUMINA , a training-free identify-then-guide framework for improved numerical alignment.

Paper PDF
O Original S Save / Search