Jiacheng Sun

papers

416

total citations

papers (11)

Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models

NEURIPS 2023arXiv

200

citations

SA-Solver: Stochastic Adams Solver for Fast Sampling of Diffusion Models

NEURIPS 2023arXiv

citations

Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels

NEURIPS 2023arXiv

citations

Towards Understanding the Generative Capability of Adversarially Robust Classifiers

ICCV 2021arXiv

citations

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

NEURIPS 2025arXiv

citations

Adding Additional Control to One-Step Diffusion with Joint Distribution Matching

ICCV 2025arXiv

citations

Masked Diffusion Models as Energy Minimization

NEURIPS 2025arXiv

citations

PolarNeXt: Rethink Instance Segmentation with Polar Representation

CVPR 2025

citations

Jiacheng Sun

papers (11)

Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models

SA-Solver: Stochastic Adams Solver for Fast Sampling of Diffusion Models

Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels

Towards Understanding the Generative Capability of Adversarially Robust Classifiers

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

Complexity Matters: Rethinking the Latent Space for Generative Modeling

Learning Few-Step Diffusion Models by Trajectory Distribution Matching

Towards Transformer-Based Aligned Generation with Self-Coherence Guidance

Adding Additional Control to One-Step Diffusion with Joint Distribution Matching

Masked Diffusion Models as Energy Minimization

PolarNeXt: Rethink Instance Segmentation with Polar Representation

papers (11)

Diff-Instruct: A Universal Approach for Transferring Knowledge From Pre-trained Diffusion Models

SA-Solver: Stochastic Adams Solver for Fast Sampling of Diffusion Models

Diffusion Models and Semi-Supervised Learners Benefit Mutually with Few Labels

Towards Understanding the Generative Capability of Adversarially Robust Classifiers

FUDOKI: Discrete Flow-based Unified Understanding and Generation via Kinetic-Optimal Velocities

Complexity Matters: Rethinking the Latent Space for Generative Modeling

Learning Few-Step Diffusion Models by Trajectory Distribution Matching

Towards Transformer-Based Aligned Generation with Self-Coherence Guidance

Adding Additional Control to One-Step Diffusion with Joint Distribution Matching

Masked Diffusion Models as Energy Minimization

PolarNeXt: Rethink Instance Segmentation with Polar Representation