Hila Chefer
Researcher at Black Forest Labs
Tel-Aviv District, Israel
Summary
🤩
Rockstar🎓
Top SchoolHila Chefer is a researcher and PhD-trained computer scientist specializing in computer vision, NLP and deep learning, currently working at Black Forest Labs after completing a direct-track PhD under Prof. Lior Wolf at Tel Aviv University. With six years of experience spanning research internships at Meta, Google and Microsoft and industry engineering at Intel, she bridges rigorous academic methods with production-aware engineering. Her open-source work includes the widely cited Transformer-Explainability PyTorch implementation (CVPR 2021), where she improved visualization and attribution methods for transformer classifiers. Comfortable across Python, deep learning frameworks and large-scale systems, she brings both theoretical insight and hands-on code refinement to model interpretability and deployment challenges.
6 years of coding experience
1 year of employment as a software developer
Bachelor of Science - BS Computer Science, Bachelor of Science - BS Computer Science at Technion - Israel Institute of TechnologyMaster of Science - MS Computer Science, Master of Science - MS Computer Science at Tel Aviv University
Hebrew, English
Github Skills (7)
vi10
transformer-models10
computer-vision10
pytorch10
explainable-artificial-intelligence10
bert9
imagenet8
Programming languages (5)
Github contributions (5)
hila-chefer/Transformer-Explainability
Nov 2020 - Jan 2023
[CVPR 2021] Official PyTorch implementation for Transformer Interpretability Beyond Attention Visualization, a novel method to visualize classifications by Transformer based networks.
Role in this project:
ML Engineer
ML EngineerContributions:82 commits, 5 PRs, 62 pushes in 2 years 1 month
Contributions summary:Hila primarily worked on improving the explainability of Transformer-based models. Their contributions involved refactoring and modifying code related to visualizing classifications, generating attention maps, and implementing different explainability methods such as LRP and rollout. The user also fixed issues related to ImageNet segmentation and made changes to the perturbation test and added examples for DeiT.
bertvisualizeattention-visualizationexplainabilityperturbation
hila-chefer/Transformer-MM-Explainability
Mar 2021 - Jan 2023
[ICCV 2021- Oral] Official PyTorch implementation for Generic Attention-model Explainability for Interpreting Bi-Modal and Encoder-Decoder Transformers, a novel method to visualize any Transformer-based network. Including examples for DETR, VQA.
Contributions:70 commits, 4 PRs, 39 pushes in 1 year 9 months
visualizeexplainabilityiccvtransformersmethod
Find and Hire Top DevelopersWe’ve analyzed the programming source code of over 60 million software developers on GitHub and scored them by 50,000 skills. Sign-up on Prog,AI to search for software developers.