Fatma

Fatma Özcan is a Principal Software Engineer at Google. Prior to that, she was a Distinguished Research Staff Member and a senior manager at IBM Almaden Research Center. Her current research focuses on platforms and infra-structure for large-scale data analysis, knowledge graphs, democratizing analytics via NLQ and conversational interfaces to data, and query processing and optimization of semi-structured data. Dr Özcan got her PhD degree in computer science from University of Maryland, College Park, and her BSc degree in computer engineering from METU, Ankara. She has over 20 years of experience in industrial research, and has delivered core technologies into many IBM products. She is the co-author of the book "Heterogeneous Agent Systems", and co-author of several conference papers and patents. She is an elected member of the SIGMOD Executive Committee, and is on the board of trustees for the VLDB Endowment and Computing Research Association. She is an ACM Distinguished Member.

Research Areas

Authored Publications
Sort By
  • Title
  • Title, descending
  • Year
  • Year, descending
Multi-Objective Agentic Rewrites for Unstructured Data Processing
Shreya Shankar
Sepanta Zeighami
Lindsey Linxi Wei
Aditya Parameswaran
Proceedings of the VLDB Endowment, VLDB Endowment (2026)
Preview abstract LLMs are being increasingly integrated into systems that support queries over unstructured data~\cite{shankar2025docetl, patel2025semantic, liu2024declarative}. In these LLM-powered data processing systems, a query is expressed as a sequence of familiar data processing operators such as map, reduce, and filter, each described in natural language for an LLM to carry out, with the system's optimizer determining the plan to execute. In this paper, we design a new optimizer to extend the rewrite directive approach to multi-objective optimization. We introduce the Multi-Objective Agentic Rewrites (MOAR) optimizer, which surfaces a Pareto frontier of pipelines that improve accuracy over the user's original while offering a range of costs---all under a limited evaluation budget (dozens of pipelines). We model the search space of pipelines (i.e., query plans) as a graph where each node is a complete pipeline and each edge applies a rewrite to produce a new one. Our insight is to apply a multi-armed bandit approach to searching the graph of rewrite sequences: at each iteration, MOAR must decide which pipeline to expand and which rewrite to apply, balancing exploration of new rewrite sequences against exploitation of promising ones. We evaluate MOAR on six real-world workloads spanning legal, medical, and enterprise domains. Compared to state-of-the-art systems and naive agentic baselines, MOAR achieves up to 1136.4\% higher output quality and 84\% lower inference cost at equivalent quality levels, while surfacing a richer set of trade-off plans for users. View details
SemBench: A Benchmark for Semantic Query Processing Engines
Jiale Lao
Gerardo Vitagliano
Immanuel Trummer
H. V. Jagadish
Sebastian Schelter
Andreas Kipf
Matthew Russo
Kris Kissel
Michael Cochez
Andreas Zimmerer
Olga Ovcharenko
Thibaud Hottelier
Gautam Gupta
Tianji Cong
2026
Preview abstract We present a benchmark targeting a novel class of systems: semantic query processing engines. Those systems rely inherently on zero-shot abilities of state-of-the-art large language models (LLMs). They extend SQL with semantic operators, configured by natural language instructions, that are evaluated via LLMs and enable users to perform various operations on multimodal data. Our benchmark provides variety along three axis: scenarios, modalities, and operators. Included are scenarios ranging from movie review analysis to medical question-answering. Within these scenarios, we cover different data modalities, including images, audio, and text. Finally, the queries involve a diverse set of operators, including semantic filters, joins, mappings, ranking, and classification operators. We evaluate systems according to processing overheads and result quality. We present experimental results for an industrial semantic query processing engine (BigQuery), as well as academic systems (LOTUS, Palimpzest, and ThalamusDB). Our results shed light on the relative strengths and weaknesses of the evaluated systems, and hint at promising avenues for future research. View details
Fine-Grained Table Retrieval for Open-Domain Tabular Question Answering
Xingyu Ji
Wojciech Kosiuk
Madelon Hulsebos
Proceedings of the 11th Workshop on Automated Knowledge Base Construction (AKBC 2026), Association for Computational Linguistics
Preview abstract This work introduces a fine-grained table retrieval framework for grounding large language models in heterogeneous, open-domain relational data. Instead of encoding a query as a single vector, the approach decomposes natural language queries into semantic components and embeds each independently, enabling more precise matching of compositional query intent. These representations are used in a staged retrieval pipeline with component-level search, connectivity-aware grouping, and reranking. Experiments on three TARGET benchmark corpora show consistent improvements in capped recall@k and stronger alignment between query intent and tabular structure over dense retrieval baselines, particularly for longer and more complex queries and when using lightweight embedding models. View details
AI Query Compilation for Unified and Optimized Execution
Proceedings of the 8th International Workshop on Applied AI for Database Systems and Applications (AIDB 2026) at VLDB (to appear)
Preview abstract In this vision paper, we propose a novel architectural paradigm for accelerated AI query execution via a unified compiled execution strategy. By compiling the hybrid \textit{AI Query} as a whole—integrating both standard SQL relational constructs and LLM inference layers into a single, unified tensor compute graph—we completely eliminate PCIe data movement bottlenecks across execution boundaries and enable global compiler optimizations. We demonstrate the viability of this unified execution paradigm on synthetic benchmarks and select AI queries from \textit{SemBench}, achieving up to 5.3x latency speedup and 9.8x throughput speedup on TPUs, and outline a research roadmap of open technical challenges to realize this vision. View details
Preview abstract SQL has been augmented with AI operators, enabling modern data analytics platforms to drive insights from both structured and unstructured data. We observe that while current Text-to-SQL systems can successfully generate these AI-augmented queries using in-context examples, the real challenge is evaluation. Traditional metrics, which rely on exact query results and deterministic execution, systematically fail against the flexible, non-deterministic outputs of embedded LLMs. In this paper, we formalize these unique evaluation failure modes and introduce a Multilayered Evaluation Framework that decouples deterministic relational database logic from flexible AI operator semantics. Using SemBench, a multimodal benchmark for SQL with AI operators, we demonstrate that traditional Execution Accuracy severely penalizes valid queries (achieving only a 36% detection rate for correct translations), even though our generator produces 70% production-viable queries. Furthermore, even state-of-the-art LLM-based auto-raters falsely reject 24% of valid queries due to the complexity of evaluating hybrid logic simultaneously. In contrast, our framework achieves a 100% detection rate on correct queries and a 91% success rate in identifying flawed translations, establishing a reliable standard for benchmarking AI-powered SQL generators. View details
High-Fidelity And Complex Test Data Generation For Real-World SQL Code Generation Services
Shivasankari Kannan
Amita Gondi
Tristan Swadell
2026 IEEE 42nd International Conference on Data Engineering (ICDE)
Preview abstract The demand for high-fidelity test data is paramount in industrial settings where access to production data is largely restricted. Traditional data generation methods often fall short, struggling with low-fidelity and the ability to model complex data structures and semantic relationships that are critical for testing complex SQL code generation services like Natural Language to SQL (NL2SQL). In this paper, we address the critical need for generating syntactically correct and semantically ``meaningful'' mock data for complex nested schema structures that are commonly expected in real-world SQL code generation services. We highlight the limitations of existing approaches used in production, particularly their inability to handle large and complex schema, as well as the lack of semantically coherent test data that lead to limited test coverage. We demonstrate that by leveraging Large Language Models (LLMs) and incorporating strategic pre- and post-processing steps, we can generate realistic high-fidelity test data that adheres to complex structural constraints and maintains semantic integrity to the test targets (SQL queries/functions). This approach supports comprehensive testing of complex SQL queries involving joins, aggregations, and even deeply nested subqueries, ensuring robust evaluation of SQL code generation services, like NL2SQL and SQL Code Assistant services. Our results demonstrate the practical utility of an out-of-the-box LLM (\textit{gemini}) based test data generation for industrial SQL code generation services where generating realistic test data is essential due to the frequent unavailability of production datasets. View details
Achieving 100x Cost & Latency Reduction: Performance Analysis of AI Query Approximation using Lightweight Proxy Models
Yannis Papakonstantinou
Xianshun Chen
Yves-Laurent Kom Samo
Yu Xiao
Jian He
Pushkar Khadilkar
Sam Idicula
Thibaud Hottelier
Alon Halevy
Rushabh Desai
Proceedings of the ACM on Management of Data, ACM New York, NY, USA (2026), pp. 1-23
Preview abstract Several data warehouse and database providers have recently introduced extensions to SQL called AI Queries, enabling users to specify conditions in SQL that are evaluated by LLMs, thereby broadening significantly the kinds of queries one can express over the combination of structured and unstructured data. LLMs offer remarkable semantic reasoning capabilities, making them an essential tool for complex and nuanced queries that blend structured and unstructured data. While extremely powerful, these AI queries can become prohibitively costly when invoked on tables with millions of rows. This paper provides an extensive evaluation of an AI query approximation approach that enables low cost analytics and database applications to benefit from AI queries. The approach deliver >100x cost and latency reduction for the semantic filter (AI.IF) operator and also important gains for semantic ranking (ORDER BY AI.SCORE). The cost and performance gains come from utilizing cheap and accurate proxy inference ML models over embedding vectors. We show that despite the massive gains in latency and cost, these proxy models preserve accuracy across various benchmark datasets, including the extended Amazon reviews benchmark that has 10M rows. We present an OLAP-friendly architecture of this approach for purely online queries, a database app-friendly architecture that further increases accuracy and the online gains by moving the proxy model training offline. We present techniques that accelerate the proxy model training. View details
Preview abstract Large Language Models (LLMs) have demonstrated impressive capabilities across a range of natural language processing tasks. In particular, improvements in reasoning abilities and the expansion of context windows have opened new avenues for leveraging these powerful models. NL2SQL is challenging in that the natural language question is inherently ambiguous, while the SQL generation requires a precise understanding of complex data schema and semantics. One approach to this semantic ambiguous problem is to provide more and sufficient contextual information. In this work, we explore the performance and the latency trade-offs of the extended context window (a.k.a., long context) offered by Google's state-of-the-art LLM (\textit{gemini-1.5-pro}). We study the impact of various contextual information, including column example values, question and SQL query pairs, user-provided hints, SQL documentation, and schema. To the best of our knowledge, this is the first work to study how the extended context window and extra contextual information can help NL2SQL generation with respect to both accuracy and latency cost. We show that long context LLMs are robust and do not get lost in the extended contextual information. Additionally, our long-context NL2SQL pipeline based on Google's \textit{gemini-pro-1.5} achieve a strong performance with 67.41\% on BIRD benchmark (dev) without finetuning and expensive self-consistency based techniques. View details
Preview abstract The integration of vector search into databases, driven by advancements in embedding models, semantic search, and Retrieval-Augmented Generation (RAG), enables powerful combined querying of structured and unstructured data. This paper focuses on filtered vector search (FVS), a core operation where relational predicates restrict the dataset before or during the vector similarity search (top-k). While approximate near neighbor (ANN) indices are commonly used to accelerate vector search by trading latency for recall, the addition of filters complicates performance optimization and makes achieving stable, declarative recall guarantees challenging. Filters alter the effective dataset size and distribution, impacting the search effort required. We discuss the primary FVS execution strategies – pre-filtering, post-filtering, and inline-filtering – whose efficiencies depend on factors like filter selectivity, cardinality, and data correlation. We review existing approaches that modify index structures and search algorithms (e.g., iterative post-filtering, filter-aware index traversal) to enhance FVS performance. This tutorial provides a comprehensive overview of filtered vector search, discussing its use cases, classifying current solutions and their trade-offs, and highlighting crucial research challenges and future directions for developing efficient and accurate FVS systems.   View details
×