Posts on DEEM Lab

SemBench accepted at VLDB

Sun, 15 Mar 2026 00:00:00 +0100

The benchmark paper SemBench: A Benchmark for Semantic Query Processing Engines has been accepted for publication at VLDB’26. This is the result of a collaboration between Google’s BigQuery team and researchers from Cornell, MIT, UTN, UMichigan, and our lab, driven by Olga from our side. Checkout sembench.org for the latest leaderboard and results!

SIGMOD Research Highlight Award for Arnab

Tue, 16 Dec 2025 00:00:00 +0100

Arnab’s 2025 paper, MEMPHIS: Holistic Lineage-based Reuse and Memory Management for Multi-backend ML Systems, has received the prestigious SIGMOD Research Highlight Award and will appear in the March 2026 special issue of SIGMOD Record. The paper has previously won the EDBT 2025 Best Research Paper Award.

DEEM at RecSys 2025

Mon, 03 Nov 2025 00:00:00 +0100

Barrie has been invited to submit an extended version of his paper on Scalable Data Debugging for Neighborhood-based Recommendation with Data Shapley Values to the Special Issue on Highlights of RecSys ’25 of the ACM Transactions on Recommender Systems.

DEEM at RecSys 2025

Fri, 26 Sep 2025 00:00:00 +0200

Barrie and Pierre are at RecSys in Prague this week! Barrie presents our work on on Scalable Data Debugging for Neighborhood-based Recommendation with Data Shapley Values, which was selected as a spotlight oral. Pierre gives a talk on his initial ideas Towards a Real-World Aligned Benchmark for Unlearning in Recommender Systems at the Responsible recommendation workshop.

DEEM at ICML 2025

Tue, 15 Jul 2025 00:00:00 +0200

Meet Olga from our lab at ICML in Vancouver this week! She will present a research paper on scSSL-Bench: Benchmarking Self-Supervised Learning for Single-Cell Data, which was selected as a spotlight poster, as well as a second paper on Towards Cross-Modal Error Detection with Tables and Images at the DataWorld workshop.

Zeyu interns with Amazon Q

Mon, 14 Jul 2025 00:00:00 +0200

Our external PhD student Zeyu from the University of Amsterdam is starting his internship with the Amazon Q team of AWS Berlin this week!

Barrie Kersbergen defended his PhD

Mon, 07 Jul 2025 00:00:00 +0200

Our external PhD Student Barrie Kersbergen (co-supervised with Maarten de Rijke) has successfully defended his PhD at the University of Amsterdam! Barrie’s research on recommender systems has been deployed to millions of users at the European e-commerce platform bol.com.

DEEM at SIGMOD 2025

Tue, 10 Jun 2025 00:00:00 +0200

Meet our lab at the SIGMOD conference in Berlin next week! We are part of the organizing committee of the conference and co-organise the DEEM workshop as well. Furthermore, we will present a workshop paper on Towards Automated Task-Aware Data Validation and run a tutorial on Navigating Data Errors in Machine Learning Pipelines on Friday.

Participation in the TRL seminar in Dagstuhl

Fri, 02 May 2025 00:00:00 +0200

Olga and Sebastian took part in the seminar on the Challenges and Opportunities of Table Representation Learning in Dagstuhl, which aims to connect the communities of data management, machine learning, and natural language processing to discuss the future of learning on tabular data.

Two talks at EDBT

Tue, 25 Mar 2025 00:00:00 +0100

Zeyu gave an invited talk about the efficient utilization of language models for table data preparation at the industry event on Next-Generation Data Management Systems at EDBT 2025 in Barcelona, and subsequently presented our paper on A Deep Dive Into Cross-Dataset Entity Matching with Large and Small Language Models.

The EU AI Act -- Developing a technical perspective

Sat, 11 Jan 2025 00:00:00 +0100

Stefan will be co-organising the workshop on Data Management for End-to-End Machine Learning (DEEM) at SIGMOD 2025 in Berlin.

The EU AI Act -- Developing a technical perspective

Tue, 03 Dec 2024 00:00:00 +0100

We have been co-organising a workshop on ‘The EU AI Act – Developing a technical perspective’ together with our colleagues from machine learning and law as well as industry practitioners.

Interview with AI Berlin

Thu, 10 Oct 2024 00:00:00 +0200

Our research group has been covered in an interview on the #ai_berlin website.

Meet us at VLDB in Chian

Sat, 01 Jun 2024 00:00:00 +0200

We will be present at the upcoming VLDB conference in China with several contributions:

How Data Management Research Helps to Improve Real World ML Applications (Keynote at the workshop on Quality in Databases)
A Flexible Forecasting Stack (Industry paper with AWS)
Snapcase - Regain Control over Your Predictions with Low-Latency Machine Unlearning (Demonstration)
Instrumentation and Analysis of Native ML Pipelines via Logical Query Plans (Paper at the PhD workshop)
Assisted design of data science pipelines (Poster of a VLDBJ publication)

Two papers at SIGMOD in Chile

Sat, 01 Jun 2024 00:00:00 +0200

At the upcoming SIGMOD conference in Chile, Till will present our paper on SchemaPile: A Large Collection of Relational Database Schemas. SchemaPile is a corpus of more than 200 thousand database schemas, which we envision to be a great resource for ML models dealing with structured data, e.g., in data integration tasks. Furthermore, Stefan will present his initial ideas on Interactively Improving ML Data Preparation Code via ‘Shadow Pipelines’ at the DEEM workshop.

Two papers at ICDE

Sun, 12 May 2024 00:00:00 +0200

Barrie and Zeyu will present two papers at the International Conference on Data Engineering (ICDE) in Utrecht. Barrie will discuss how to choose cost-efficient deployment options for neural recommendation models in e-commerce, while Zeyu will present initial ideas for zero-shot entity matching.

Poster at ICLR

Mon, 06 May 2024 00:00:00 +0200

The ‘DataScope’ paper on debugging ML pipelines via Shapley importance will be presented at the International Conference on Learning Representations (ICLR) in Vienna. This work was driven by Bojan Karlaš from Harvard and Ce Zhang from the University of Chicago.

Action Editor at DMLR

Wed, 18 Oct 2023 00:00:00 +0200

Prof. Schelter joins the newly formed Journal of Data-centric Machine Learning Research (DMLR) as an Action Editor.