3rd Workshop on Efficient and On-Device Generation (EDGE), CVPR 2026

On the Limits of Token Reduction for Efficient Unified Vision Language Training

Siyi Chen,

Weiming Zhuang,

Jingtao Li,

Lingjuan Lv

[pdf]

[bibtex]

@InProceedings{Chen_2026_CVPR, author = {Chen, Siyi and Zhuang, Weiming and Li, Jingtao and Lv, Lingjuan}, title = {On the Limits of Token Reduction for Efficient Unified Vision Language Training}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3400-3409} }

Training-Free Layout-to-Image Generation with Marginal Attention Constraints

Huancheng Chen,

Jingtao Li,

Weiming Zhuang,

Haris Vikalo,

Lingjuan Lyu

[pdf] [supp] [arXiv]

[bibtex]

@InProceedings{Chen_2026_CVPR, author = {Chen, Huancheng and Li, Jingtao and Zhuang, Weiming and Vikalo, Haris and Lyu, Lingjuan}, title = {Training-Free Layout-to-Image Generation with Marginal Attention Constraints}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3410-3419} }

CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers

Zhuojin Li,

Hsin-Pai Cheng,

Hong Cai,

Shizhong Han,

Fatih Porikli

[pdf] [arXiv]

[bibtex]

@InProceedings{Li_2026_CVPR, author = {Li, Zhuojin and Cheng, Hsin-Pai and Cai, Hong and Han, Shizhong and Porikli, Fatih}, title = {CoReDiT: Spatial Coherence-Guided Token Pruning and Reconstruction for Efficient Diffusion Transformers}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3438-3447} }

A KL Lens on Quantization: Fast, Forward-Only Sensitivity for Mixed-Precision SSM-Transformer Models

Jason Kong,

Nilesh Prasad Pandey,

Flavio Ponzina,

Tajana Rosing

[pdf] [supp] [arXiv]

[bibtex]

@InProceedings{Kong_2026_CVPR, author = {Kong, Jason and Pandey, Nilesh Prasad and Ponzina, Flavio and Rosing, Tajana}, title = {A KL Lens on Quantization: Fast, Forward-Only Sensitivity for Mixed-Precision SSM-Transformer Models}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3429-3437} }

ObjectAlign: Neuro-Symbolic Object Consistency Verification and Correction

Mustafa Munir,

Harsh Goel,

Xiwen Wei,

Minkyu Choi,

Sahil Shah,

Kartikeya Bhardwaj,

Paul Whatmough,

Sandeep Chinchali,

Radu Marculescu

[pdf] [arXiv]

[bibtex]

@InProceedings{Munir_2026_CVPR, author = {Munir, Mustafa and Goel, Harsh and Wei, Xiwen and Choi, Minkyu and Shah, Sahil and Bhardwaj, Kartikeya and Whatmough, Paul and Chinchali, Sandeep and Marculescu, Radu}, title = {ObjectAlign: Neuro-Symbolic Object Consistency Verification and Correction}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3458-3468} }

Token-Aligned Hierarchies for Lightweight Super-Resolution

Jinendra Malekar,

Lingjia Shi,

Peyton Chandarana,

Ramtin Zand

[pdf]

[bibtex]

@InProceedings{Malekar_2026_CVPR, author = {Malekar, Jinendra and Shi, Lingjia and Chandarana, Peyton and Zand, Ramtin}, title = {Token-Aligned Hierarchies for Lightweight Super-Resolution}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3448-3457} }

DirectFlow: One-Step Posterior Transport Distillation for Blind Face Restoration

Eashan Adhikarla,

Brian D Davison

[pdf]

[bibtex]

@InProceedings{Adhikarla_2026_CVPR, author = {Adhikarla, Eashan and Davison, Brian D}, title = {DirectFlow: One-Step Posterior Transport Distillation for Blind Face Restoration}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3389-3399} }

Diagonal-Tiled Mixed-Precision Attention for Efficient Low-Bit MXFP Inference

Yifu Ding,

Xinhao Zhang,

Yinglei Wang,

Yuanpei Chen,

Jinyang Guo

[pdf]

[bibtex]

@InProceedings{Ding_2026_CVPR, author = {Ding, Yifu and Zhang, Xinhao and Wang, Yinglei and Chen, Yuanpei and Guo, Jinyang}, title = {Diagonal-Tiled Mixed-Precision Attention for Efficient Low-Bit MXFP Inference}, booktitle = {Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) Workshops}, month = {June}, year = {2026}, pages = {3420-3428} }