Immanuel Trummer
Associate Professor, Cornell University
Immanuel Trummer is an associate professor of computer science at Cornell University. His research makes data analysis more efficient and more user-friendly, in particular by the use of large language models. He received the NSF CAREER grant, and several "Best of VLDB" and "Best of SIGMOD" awards. He is also the author of the book "Data Analysis with LLMs", now available in five languages.
Publications
2026
Data Engineering Bulletin Optimal Block Nested Loops Implementations for Semantic Joins.
Data Engineering Bulletin ThalamusDB: Semantic Approximate Query Processing.
Data Engineering Bulletin Towards Token-Efficient Query Languages for Data Agents.
PVLDB SemBench: A Benchmark for Semantic Query Processing Engines.
PVLDB Hybrid Mixed Integer Linear Programming for Large-Scale Join Order Optimisation.
PVLDB Demonstrating TACT: Tunable Accuracy-Cost Trade-offs for Semantic Image Filtering.
Q-Data Leveraging Quantum Annealing for Materialized View Selection.
Q-Data Quantum Annealing for Multiobjective Query Optimization.
SIGMOD Large-Scale Multiple Query Optimisation with Incremental Quantum(-Inspired) Annealing.
2025
CDMS Rethinking Pluggable Federated Query Optimization: From Laptops to Data Warehouse.
Companion of the 2025 International Conference on Management of Data SwellDB: Dynamic Query-Driven Table Generation with Large Language Models.
Data Engineering Bulletin Customizing Operator Implementations for SQL Processing via Large Language Models.
Manning Publications Data Analysis with LLMs: Text, Tables, Images and Sound.
Proceedings of the VLDB Endowment CEDAR: A System for Cost-Efficient Data-Driven Claim Verification.
Q-Data Cost-Based Query Optimization for Quantum Computation.
Q-Data Leveraging Quantum Computing for Optimal Data Allocation in Distributed Systems.
SIGMOD Demonstrating CEDAR: A System for Cost-Efficient Data-Driven Claim Verification.
SIGMOD Lambda-Tune: Harnessing Large Language Models for Automated Database System Tuning.
SIGMOD Companion Second Workshop on Quantum Computing and Quantum-Inspired Technology for Data-Intensive Systems and Applications (Q-Data).
VLDBJ Generating Highly Customizable Python Code for Data Processing with Large Language Models. Best of VLDB
2024
PVLDB DBG-PT: A Large Language Model Assisted Query Performance Regression Debugger.
Q-Data Towards Out-of-Core Simulators for Quantum Computing.
Q-Data Leveraging Quantum Computing for Database Index Selection.
SIGMOD ROME: Robust Query Optimization via Parallel Multi-Plan Execution.
SIGMOD ThalamusDB: Approximate Query Processing on Multi-Modal Data.
SIGMOD Demonstrating λ-Tune: Exploiting Large Language Models for Workload-Adaptive Database System Tuning.
VLDB Journal DB-BERT: Making Database Tuning Tools “Read” the Manual.
2023
CoRR Language Models Enable Simple Systems for Generating Structured Views of Heterogeneous Data Lakes.
PVLDB Can Large Language Models Predict Data Correlations from Column Names?
PVLDB Demonstrating GPT-DB: Generating Query-Specific and Customizable Code for SQL Processing with GPT-4.
SIGMOD Demonstrating NaturalMiner: Searching Large Data Sets for Abstract Patterns Described in Natural Language. Best Demo Runner Up
2022
AAAI Procrastinated Tree Search: Black-box Optimization with Delayed, Noisy, and Multi-fidelity Feedback.
PVLDB BABOONS: Black-Box Optimization of Data Summaries in Natural Language.
PVLDB BABOONS: Black-box optimization of data summaries in natural language.
PVLDB Synthesizing Code for Query Processing from Natural Language Instructions using GPT-3 Codex.
PVLDB CodexDB: Synthesizing Code for Query Processing from Natural Language Instructions using GPT-3 Codex.
PVLDB From BERT to GPT-3 Codex: Harnessing the Potential of Very Large Language Models for Data Management.
SIGMOD DB-BERT: a Database Tuning Tool that ``Reads the Manual''.
SIGMOD Demonstrating DB-BERT: A Database Tuning Tool that "Reads" the Manual.
VLDB PhD Workshop Building Learned Federated Query Optimizers.
2021
ACM Transactions on Database Systems SkinnerDB: Regret-bounded Query Evaluation via Reinforcement Learning. Best of SIGMOD
https://arxiv.org/pdf/2107.04553.pdf Can deep neural networks predict data correlations from column names?
ICDE Optimally Summarizing Data by Small Fact Sets for Concise Answers to Voice Queries.
IEEE Data Eng. Bull. WebChecker: Towards an Infrastructure for Efficient Misinformation Detection at Web Scale.
PVLDB The Case for NLP-Enhanced Database Tuning: Towards Tuning Tools that “Read the Manual”.
PVLDB UDO: Universal Database Optimization using Reinforcement Learning.
PVLDB Robust Voice Querying with MUVE: Optimally Visualizing Results of Phonetically Similar Queries.
SIGMOD Record Database Tuning using Natural Language Processing.
2020
BDA Scrutinizer: a System for Checking Statistical Claims. Best Demonstration Award
CIDR BitGourmet: deterministic approximation via optimized bit selection.
CIDR BitGourmet: Deterministic Approximation via Optimized Bit Selections.
PVLDB Demonstrating the Voice-Based Exploration of Large Data Sets with CiceroDB-Zero.
PVLDB Scrutinizer: A Mixed-Initiative Approach to Large-Scale, Data-Driven Claim Verification.
SIGMOD Demonstration of BitGourmet: Data Analysis via Deterministic Approximation.
2019
PVLDB AggChecker: a Fact-Checking System for Text Summaries of Relational Data Sets.
SIGMOD Exact Cardinality Query Optimization with Bounded Execution Cost.
SIGMOD A Holistic Approach for Query Evaluation and Result Vocalization in Voice-Based OLAP.
SIGMOD SkinnerDB: Regret-Bounded Query Evaluation via Reinforcement Learning. Selected for Best of SIGMOD
2018
2017
Communications of the ACM Multi-Objective Parametric Query Optimization.
PVLDB Data Vocalization: Optimizing Voice Output of Relational Data.
SIGMOD Solving the Join Ordering Problem via Mixed Integer Linear Programming.
Funding
External
- FDA Grant: SAGE: An Integrated Monitoring System for Antimicrobial Use and Stewardship (2027–2031).
- LinkedIn Grant: Making Agentic Data Analysis Token-Efficient (2026–2027).
- NSF 2239326: CAREER: Mining Hints from Text Documents to Guide Automated Database Performance Tuning (2023–2028).
- NSF 1910830: III: Small: Regret-Bounded Query Evaluation via Reinforcement Learning (2019–2022).
- Google AI-Powered Immediate Response to Pandemics Initiative: Data-Driven Automatic Fact Checking of Coronavirus Claims (2020).
- Fact-Checking Development Grant (IFCN & YouTube): Automatic Fact-Checking of Coronavirus Claims (2020).
- Google Faculty Research Award: Mining an "Anti-Knowledge Base" for Fact Checking from Wikipedia Updates (2018).
- Google Faculty Research Award: Optimizing Voice-Based Output of Relational Data (2016).
Internal
- CCSS Grant: Towards a National Infrastructure of Federated Private Computation Spaces (2026–2027).
- CIDA RIF Seed Grant: Co-Creating a Human-Machine Interface Better Adapted for On-Farm Data Recording, Curation, Management, and Use (2021–2023).
- Huawei Research Initiative: Deterministic Approximation via BitGourmet (2019).
Students
Postdoctoral researchers
- Manuel Schönberger (website)
PhD students
Graduated PhD students
First position listed is where each student went after completing the PhD.
Teaching
Cornell Courses
- CS 4320: Introduction to database systems.
- CS 4321: Practicum in database systems.
- CS 5322: Efficient analysis of large data sets.
- CS 6320: Advanced database systems.
- CS 7390: Seminar in database systems.
Online Courses
- Introductory database lecture (over 1M views!)
- Tutorial on large language models