✨Test-time scaling of diffusions with flow maps
📝 Summary:
The Flow Map Trajectory Tilting FMTT algorithm enhances test-time diffusion models by using flow maps to align better with user rewards. This approach solves the ill-posed problem of reward gradients, achieving superior reward ascent for improved sampling and novel image editing.
🔹 Publication Date: Published on Nov 27
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.22688
• PDF: https://arxiv.org/pdf/2511.22688
• Project Page: https://flow-map-trajectory-tilting.github.io/
==================================
For more data science resources:
✓ https://t.me/DataScienceT
#DiffusionModels #GenerativeAI #ImageEditing #MachineLearning #FlowMaps
📝 Summary:
The Flow Map Trajectory Tilting FMTT algorithm enhances test-time diffusion models by using flow maps to align better with user rewards. This approach solves the ill-posed problem of reward gradients, achieving superior reward ascent for improved sampling and novel image editing.
🔹 Publication Date: Published on Nov 27
🔹 Paper Links:
• arXiv Page: https://arxiv.org/abs/2511.22688
• PDF: https://arxiv.org/pdf/2511.22688
• Project Page: https://flow-map-trajectory-tilting.github.io/
==================================
For more data science resources:
✓ https://t.me/DataScienceT
#DiffusionModels #GenerativeAI #ImageEditing #MachineLearning #FlowMaps
❤1