Olga Ovcharenko
@o_ovcharenko
Followers
28
Following
229
Media
7
Statuses
23
Ph.D. at @bifoldberlin | M. Sc. Data Science student at @ETH
Joined July 2020
Cross-modal errors span across multiple modaltities, like mismatched images or incorrect text fields, and are are an overlooked but critical issue in domains like e-Commerce and healthcare. Even with modern tools, large orgs struggle to catch cross-modal errors.
1
0
2
✨ Excited to present our workshop paper at DataWorld at #ICML2025 tomorrow 🇨🇦. We introduce the problem of detecting cross-modal errors in tabular data that originate from other modalities. Visit our poster:.📅 Saturday, July 19, 10:05 AM - 11:20 AM.📍 West Meeting Room 208-209
1
1
5
Thanks to all co-authors Florian Barkmann, Philip Toma, @ImantDaunhawer, @vogt_je, @sscdotopen and.@val_boeva. 📄 Full paper: 💻 Code:
github.com
Contribute to BoevaLab/scSSL-Bench development by creating an account on GitHub.
0
1
5
📊 Key findings:.- Specialized single-cell frameworks and scFMs excel at uni-modal batch correction.- Generic SSL methods like VICReg & SimCLR outperform on cell typing & multi-modal tasks.- Masking emerges as the most universally effective augmentation strategy.
1
0
2
RT @val_boeva: Join our lab's presentations at ICML'2025 @icmlconf in beautiful Vancouver!. 1. Thursday, Olga Ovcharenko (@o_ovcharenko) wi….
0
2
0
RT @sscdotopen: The DEEM Lab is at ICML this week for the first time, with two contributions! . (1/3).
0
2
0
Our paper "Towards Cross-Modal Error Detection with Tables and Images" was accepted for the DataWorld workshop at ICML'25! 🥳 . Thanks to @sscdotopen!
1
1
11
RT @sscdotopen: We have a PhD opening in Berlin on "Responsible Data Engineering", with a focus on data preparation pipelines designed alon….
0
5
0
RT @sscdotopen: We have a PhD opening in Berlin on "Responsible Data Engineering", with a focus on data preparation for ML/AI systems. Thi….
0
5
0
RT @ImmanuelTrummer: Had a great time at the @dagstuhl seminar on Table Representation Learning! Lots of interesting discussions and future….
0
1
0
📢 Our extended benchmark on self-supervised learning for single-cell data, scSSL-Bench 🧬, is now accepted at ICML (spotlight)!. Thanks to all collaborators from @bifoldberlin and @CSatETH !.
📢Our benchmark on self-supervised learning for single-cell data🧬 is accepted at the #NeurIPS2024 SSL workshop. We take a first step towards establishing best practices for SSL methods for single-cell data, and benchmark 8 SSL methods on 3 downstream tasks across 8 datasets.
0
5
13
RT @sscdotopen: We have openings for student assistants in the DEEM Lab at @bifoldberlin . This is a great opportunity to work with PhD stu….
0
9
0
RT @SGrafberger: Our vision paper "Towards Regaining Control over Messy ML Pipelines" was accepted for the @DAIS_workshop at @icdeconf!. In….
0
2
0
📄Read the full preprint: Thanks to all co-authors Philip Toma, @ImantDaunhawer, @vogt_je, @flobarkmann, and @val_boeva.
0
0
4
Furthermore, we find that masking leads to the biggest performance improvement for creating augmented views of a cell.
1
0
2
Our bechmark shows that SimCLR and VICReg are the best methods for single-cell data across batch integration, query-to-reference, and missing modality prediction tasks.
1
0
1
📢Our benchmark on self-supervised learning for single-cell data🧬 is accepted at the #NeurIPS2024 SSL workshop. We take a first step towards establishing best practices for SSL methods for single-cell data, and benchmark 8 SSL methods on 3 downstream tasks across 8 datasets.
1
2
14
