Hila Chefer

Hi there! I am a PhD Candidate at Tel Aviv University, working in the Deep Learning Lab under the supervision of Prof. Lior Wolf. Also, I am currently a research intern at Meta AI in Tel Aviv. Before that, I was a research intern at Google. My research is centered around computer vision and multi-modal learning. I am particularly passionate about developing tools to enhance interpretability, reliability and controllability of deep foundation models, using the model's internal representations. My work has been covered by The Verge, ZDNET, Analytics India Magazine, and others.

Email  /  Google Scholar  /  Twitter  /  GitHub  /  Semantic Scholar  /  LinkedIn   

profile photo

Selected Publications

* indicates equal contribution.

Still-Moving: Customized Video Generation without Customized Video Data
Hila Chefer, Shiran Zada, Roni Paiss, Ariel Efrat, Omer Tov, Michael Rubinstein , Lior Wolf, Tali Dekel, Tomer Michaeli, Inbar Mosseri
SIGGRAPH Asia (Journal), 2024
Project page / Paper

Lumiere: A Space-Time Diffusion Model for Video Generation
Omer Bartal*, Hila Chefer*, Omer Tov*, Charles Herrmann, Roni Paiss, Shiran Zada, Ariel Efrat, Junhwa Hur , Yuanzhen Li Tomer Michaeli, Oliver Wang, Deqing Sun, Tali Dekel, Inbar Mosseri
SIGGRAPH Asia, 2024
Project page / Paper

The Hidden Language of Diffusion Models
Hila Chefer, Oran Lang, Mor Geva, Volodymyr Polosukhin,
Assaf Shocher, Michal Irani, Inbar Mosseri, Lior Wolf
ICLR, 2024
Project page / Paper

Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-Image Diffusion Models
Hila Chefer*, Yuval Alaluf*, Yael Vinker, Lior Wolf, Daniel Cohen-Or
SIGGRAPH (journal), 2023
Project page / Paper

Optimizing Relevance Maps of Vision Transformers Improves Robustness
Hila Chefer, Idan Schwartz, Lior Wolf
NeurIPS, 2022
Project page / Paper

Image-Based Clip-Guided Essence Transfer
Hila Chefer, Sagie Benaim, Roni Paiss, Lior Wolf
ECCV, 2022
Project page / Paper

No Token Left Behind: Explainability-Aided Image Classification and Generation
Roni Paiss, Hila Chefer, Lior Wolf
ECCV, 2022
Paper

Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers
Hila Chefer, Shir Gur, Lior Wolf
ICCV, 2021 (Oral)
Project page / Paper

Transformer Interpretability Beyond Attention Visualization
Hila Chefer, Shir Gur, Lior Wolf
CVPR, 2021
Project page / Paper

Selected Talks

.
All Things ViTs: Understanding and Interpreting Attention in Vision
Hila Chefer*, Sayak Paul*
CVPR Tutorial, 2023
Tutorial page / Recording

Transformer Explainability Beyond Accountability
Hila Chefer
Columbia Vision Seminar
Recording

Intro to Transformers and Transformer Explainability
Hila Chefer
Microsoft Data Science Bond
Recording

Lumiere: A Space-Time Diffusion Model for Video Generation
Hila Chefer
Recording

Leveraging Attention for Improved Accuracy and Robustness
Hila Chefer
Voxel51 Computer Vision Meetup
Recording

Attend-and-Excite: Attention-Based Semantic Guidance for Text-to-image Diffusion Models
Hila Chefer
Microsoft Data Science Bond
Recording

Contact: hilach70 at gmail dot com


webpage template from Jon Barron