影响力指数

87.11/100

前 0.8%

全站排名 #504

发表论文32 篇

平均评分5.8

年均产出10.7 篇/年

Sergey Tulyakov

Director of Research@Snap Inc.·美国·OpenReview

研究方向

Image synthesis · manipulation style transfer · Video syntesis · manipulation · prediction · retargeting · Generative models · Face analysis

6.0

H3AE: High Compression, High Speed, and High Quality AutoEncoder for Video Diffusion Models

ICLR 2026Rejected

通讯

4.0

LayerComposer: Interactive Personalized T2I via Spatially-Aware Layered Canvas

ICLR 2026Withdrawn

-1

One Model, Many Budgets: Elastic Latent Interfaces for Diffusion Transformers

ICLR 2026Desk Rejected

7.8

Preventing Shortcuts in Adapter Training via Providing the Shortcuts

NeurIPS 2025Poster

7.3

Fused View-Time Attention and Feedforward Reconstruction for 4D Scene Generation

NeurIPS 2025Poster

7.0

Lightweight Predictive 3D Gaussian Splats

ICLR 2025Poster

6.8

Improving Progressive Generation with Decomposable Flow Matching

NeurIPS 2025Poster

6.8

Towards Physical Understanding in Video Generation: A 3D Point Regularization Approach

NeurIPS 2025Poster

6.8

DenseDPO: Fine-Grained Temporal Preference Optimization for Video Diffusion Models

NeurIPS 2025Spotlight

6.6

Improving the Diffusability of Autoencoders

ICML 2025Poster

6.2

VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control

ICLR 2025Poster

通讯

6.1

I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models

ICML 2025Poster

6.0

Scalable Ranked Preference Optimization for Text-to-Image Generation

ICLR 2025Rejected

6.0

DELTA: DENSE EFFICIENT LONG-RANGE 3D TRACKING FOR ANY VIDEO

ICLR 2025Poster

5.8

ControlMM: Controllable Masked Motion Generation

ICLR 2025Rejected

通讯

5.6

GTR: Improving Large 3D Reconstruction Models through Geometry and Texture Refinement

ICLR 2025Poster

5.3

Taming Data and Transformers for Audio Generation

ICLR 2025Rejected

4.8

VideoAlchemy: Open-set Personalization in Video Generation

ICLR 2025Withdrawn

通讯

4.7

VIA: Unified Spatiotemporal Video Adaptation for Global and Local Video Editing

合作者 (20)

Sergey Tulyakov

AlphaFlow: Understanding and Improving MeanFlow Models

SPRINT: Sparse-Dense Residual Fusion for Efficient Diffusion Transformers

ShapeGen4D: Towards High Quality 4D Shape Generation from Videos

Taming Diffusion Transformer for Efficient Mobile Video Generation in Seconds

H3AE: High Compression, High Speed, and High Quality AutoEncoder for Video Diffusion Models

LayerComposer: Interactive Personalized T2I via Spatially-Aware Layered Canvas

One Model, Many Budgets: Elastic Latent Interfaces for Diffusion Transformers

Preventing Shortcuts in Adapter Training via Providing the Shortcuts

Fused View-Time Attention and Feedforward Reconstruction for 4D Scene Generation

Lightweight Predictive 3D Gaussian Splats

Improving Progressive Generation with Decomposable Flow Matching

Towards Physical Understanding in Video Generation: A 3D Point Regularization Approach

DenseDPO: Fine-Grained Temporal Preference Optimization for Video Diffusion Models

Improving the Diffusability of Autoencoders

VD3D: Taming Large Video Diffusion Transformers for 3D Camera Control

I Think, Therefore I Diffuse: Enabling Multimodal In-Context Reasoning in Diffusion Models

Scalable Ranked Preference Optimization for Text-to-Image Generation

DELTA: DENSE EFFICIENT LONG-RANGE 3D TRACKING FOR ANY VIDEO

ControlMM: Controllable Masked Motion Generation

GTR: Improving Large 3D Reconstruction Models through Geometry and Texture Refinement

Taming Data and Transformers for Audio Generation

VideoAlchemy: Open-set Personalization in Video Generation

VIA: Unified Spatiotemporal Video Adaptation for Global and Local Video Editing