arXiv subjects

Chuan Lei

Publications and source records attributed to Chuan Lei.

At least 19 recordsLinked to original sources

arXiv2026

SOMA-SQL: Resolving Multi-Source Ambiguity in NL-to-SQL via Synthetic Log and Execution Probing

Natural language interfaces to databases aim to translate user questions into executable SQL, yet remain brittle in real-world settings where questions are underspecified and schemas are large and ambiguous. Ambiguity across user questions, database schemas, and model interpretations are central failure modes in NL2SQL, leading to misaligned intent, incorrect schema grounding, and erroneous SQL generation. Existing approaches rely on human clarification or treat ambiguity as a schema representation problem, but these do not scale nor resolve ambiguity autonomously. We propose SOMA-SQL to automatically resolve ambiguity via targeted synthetic query log and ambiguity-driven probing. SOMA-SQL constructs synthetic query log to ground schema interpretation and guide candidate SQL generation; it then executes targeted probing queries, driven by a structured ambiguity taxonomy and candidate disagreements, to produce disambiguation evidence for final SQL selection and repair. This active approach to ambiguity discovery and resolution generalizes across unseen schemas and query distributions without human-in-the-loop. Experiments on six public benchmarks demonstrate that SOMA-SQL improves execution accuracy by 13.0% on average over state-of-the-art baselines, with gains of up to 16.7% on ambiguous questions.

cs.CL↗

arXiv2026

AgentSM: Semantic Memory for Agentic Text-to-SQL

Recent advances in LLM-based Text-to-SQL have achieved remarkable gains on public benchmarks such as BIRD and Spider. Yet, these systems struggle to scale in realistic enterprise settings with large, complex schemas, diverse SQL dialects, and expensive multi-step reasoning. Emerging agentic approaches show potential for adaptive reasoning but often suffer from inefficiency and instability-repeating interactions with databases, producing inconsistent outputs, and occasionally failing to generate valid answers. To address these challenges, we introduce Agent Semantic Memory (AgentSM), an agentic framework for Text-to-SQL that builds and leverages interpretable semantic memory. Instead of relying on raw scratchpads or vector retrieval, AgentSM captures prior execution traces-or synthesizes curated ones-as structured programs that directly guide future reasoning. This design enables systematic reuse of reasoning paths, which allows agents to scale to larger schemas, more complex questions, and longer trajectories efficiently and reliably. Compared to state-of-the-art systems, AgentSM achieves higher efficiency by reducing average token usage and trajectory length by 25% and 35%, respectively, on the Spider 2.0 benchmark. It also improves execution accuracy, reaching a state-of-the-art accuracy of 44.8% on the Spider 2.0 Lite benchmark.

cs.AI↗

arXiv2025

LAKEGEN: A LLM-based Tabular Corpus Generator for Evaluating Dataset Discovery in Data Lakes

How to generate a large, realistic set of tables along with joinability relationships, to stress-test dataset discovery methods? Dataset discovery methods aim to automatically identify related data assets in a data lake. The development and evaluation of such solutions for customers from a wide range of business domains, relies on diverse, high quality and domain-specific tabular benchmarks. Large language models (LLMs) are trained on a wide variety of text data, which can provide a strong foundation of general and domain-specific knowledge. In this paper, we ask the question -- \textit{can we leverage LLMs to generate a tabular benchmark adequate for evaluating the dataset discovery solutions?} In particular, we focus on the task of finding joinable tables which is the cornerstone of virtually every dataset discovery method. Current corpora for evaluating dataset discovery methods are mainly based on subsets of open data, and they suffer from three important issues: $i)$ they focus on very common and generic data types (e.g., address, id, name, etc.); $ii)$ they do not contain human-annotated column pairs; instead, practitioners synthesize ground truth using table splits (e.g., horizontal for table union search and vertical ones for joinability) and $iii)$ they do not focus on semantic column relationships.

cs.DB↗

arXiv2025

SQLens: An End-to-End Framework for Error Detection and Correction in Text-to-SQL

Text-to-SQL systems translate natural language (NL) questions into SQL queries, enabling non-technical users to interact with structured data. While large language models (LLMs) have shown promising results on the text-to-SQL task, they often produce semantically incorrect yet syntactically valid queries, with limited insight into their reliability. We propose SQLens, an end-to-end framework for fine-grained detection and correction of semantic errors in LLM-generated SQL. SQLens integrates error signals from both the underlying database and the LLM to identify potential semantic errors within SQL clauses. It further leverages these signals to guide query correction. Empirical results on two public benchmarks show that SQLens outperforms the best LLM-based self-evaluation method by 25.78% in F1 for error detection, and improves execution accuracy of out-of-the-box text-to-SQL systems by up to 20%.

cs.CL↗

arXiv2025

TailorSQL: An NL2SQL System Tailored to Your Query Workload

NL2SQL (natural language to SQL) translates natural language questions into SQL queries, thereby making structured data accessible to non-technical users, serving as the foundation for intelligent data applications. State-of-the-art NL2SQL techniques typically perform translation by retrieving database-specific information, such as the database schema, and invoking a pre-trained large language model (LLM) using the question and retrieved information to generate the SQL query. However, existing NL2SQL techniques miss a key opportunity which is present in real-world settings: NL2SQL is typically applied on existing databases which have already served many SQL queries in the past. The past query workload implicitly contains information which is helpful for accurate NL2SQL translation and is not apparent from the database schema alone, such as common join paths and the semantics of obscurely-named tables and columns. We introduce TailorSQL, a NL2SQL system that takes advantage of information in the past query workload to improve both the accuracy and latency of translating natural language questions into SQL. By specializing to a given workload, TailorSQL achieves up to 2$\times$ improvement in execution accuracy on standardized benchmarks.

cs.DB↗

arXiv2025

ODIN: A NL2SQL Recommender to Handle Schema Ambiguity

NL2SQL (natural language to SQL) systems translate natural language into SQL queries, allowing users with no technical background to interact with databases and create tools like reports or visualizations. While recent advancements in large language models (LLMs) have significantly improved NL2SQL accuracy, schema ambiguity remains a major challenge in enterprise environments with complex schemas, where multiple tables and columns with semantically similar names often co-exist. To address schema ambiguity, we introduce ODIN, a NL2SQL recommendation engine. Instead of producing a single SQL query given a natural language question, ODIN generates a set of potential SQL queries by accounting for different interpretations of ambiguous schema components. ODIN dynamically adjusts the number of suggestions based on the level of ambiguity, and ODIN learns from user feedback to personalize future SQL query recommendations. Our evaluation shows that ODIN improves the likelihood of generating the correct SQL query by 1.5-2$\times$ compared to baselines.

cs.DB↗

arXiv2024

FeatNavigator: Automatic Feature Augmentation on Tabular Data

Data-centric AI focuses on understanding and utilizing high-quality, relevant data in training machine learning (ML) models, thereby increasing the likelihood of producing accurate and useful results. Automatic feature augmentation, aiming to augment the initial base table with useful features from other tables, is critical in data preparation as it improves model performance, robustness, and generalizability. While recent works have investigated automatic feature augmentation, most of them have limited capabilities in utilizing all useful features as many of them are in candidate tables not directly joinable with the base table. Worse yet, with numerous join paths leading to these distant features, existing solutions fail to fully exploit them within a reasonable compute budget. We present FeatNavigator, an effective and efficient framework that explores and integrates high-quality features in relational tables for ML models. FeatNavigator evaluates a feature from two aspects: (1) the intrinsic value of a feature towards an ML task (i.e., feature importance) and (2) the efficacy of a join path connecting the feature to the base table (i.e., integration quality). FeatNavigator strategically selects a small set of available features and their corresponding join paths to train a feature importance estimation model and an integration quality prediction model. Furthermore, FeatNavigator's search algorithm exploits both estimated feature importance and integration quality to identify the optimized feature augmentation plan. Our experimental results show that FeatNavigator outperforms state-of-the-art solutions on five public datasets by up to 40.1% in ML model performance.

cs.DB↗

arXiv2024

4DBInfer: A 4D Benchmarking Toolbox for Graph-Centric Predictive Modeling on Relational DBs

Although RDBs store vast amounts of rich, informative data spread across interconnected tables, the progress of predictive machine learning models as applied to such tasks arguably falls well behind advances in other domains such as computer vision or natural language processing. This deficit stems, at least in part, from the lack of established/public RDB benchmarks as needed for training and evaluation purposes. As a result, related model development thus far often defaults to tabular approaches trained on ubiquitous single-table benchmarks, or on the relational side, graph-based alternatives such as GNNs applied to a completely different set of graph datasets devoid of tabular characteristics. To more precisely target RDBs lying at the nexus of these two complementary regimes, we explore a broad class of baseline models predicated on: (i) converting multi-table datasets into graphs using various strategies equipped with efficient subsampling, while preserving tabular characteristics; and (ii) trainable models with well-matched inductive biases that output predictions based on these input subgraphs. Then, to address the dearth of suitable public benchmarks and reduce siloed comparisons, we assemble a diverse collection of (i) large-scale RDB datasets and (ii) coincident predictive tasks. From a delivery standpoint, we operationalize the above four dimensions (4D) of exploration within a unified, scalable open-source toolbox called 4DBInfer. We conclude by presenting evaluations using 4DBInfer, the results of which highlight the importance of considering each such dimension in the design of RDB predictive models, as well as the limitations of more naive approaches such as simply joining adjacent tables. Our source code is released at https://github.com/awslabs/multi-table-benchmark .

cs.LG↗

arXiv2024

OpenTab: Advancing Large Language Models as Open-domain Table Reasoners

Large Language Models (LLMs) trained on large volumes of data excel at various natural language tasks, but they cannot handle tasks requiring knowledge that has not been trained on previously. One solution is to use a retriever that fetches relevant information to expand LLM's knowledge scope. However, existing textual-oriented retrieval-based LLMs are not ideal on structured table data due to diversified data modalities and large table sizes. In this work, we propose OpenTab, an open-domain table reasoning framework powered by LLMs. Overall, OpenTab leverages table retriever to fetch relevant tables and then generates SQL programs to parse the retrieved tables efficiently. Utilizing the intermediate data derived from the SQL executions, it conducts grounded inference to produce accurate response. Extensive experimental evaluation shows that OpenTab significantly outperforms baselines in both open- and closed-domain settings, achieving up to 21.5% higher accuracy. We further run ablation studies to validate the efficacy of our proposed designs of the system.

cs.LG↗

arXiv2024

OmniMatch: Effective Self-Supervised Any-Join Discovery in Tabular Data Repositories

How can we discover join relationships among columns of tabular data in a data repository? Can this be done effectively when metadata is missing? Traditional column matching works mainly rely on similarity measures based on exact value overlaps, hence missing important semantics or failing to handle noise in the data. At the same time, recent dataset discovery methods focusing on deep table representation learning techniques, do not take into consideration the rich set of column similarity signals found in prior matching and discovery methods. Finally, existing methods heavily depend on user-provided similarity thresholds, hindering their deployability in real-world settings. In this paper, we propose OmniMatch, a novel join discovery technique that detects equi-joins and fuzzy-joins betwen columns by combining column-pair similarity measures with Graph Neural Networks (GNNs). OmniMatch's GNN can capture column relatedness leveraging graph transitivity, significantly improving the recall of join discovery tasks. At the same time, OmniMatch also increases the precision by augmenting its training data with negative column join examples through an automated negative example generation process. Most importantly, compared to the state-of-the-art matching and discovery methods, OmniMatch exhibits up to 14% higher effectiveness in F1 score and AUC without relying on metadata or user-provided thresholds for each similarity metric.

cs.DB↗

arXiv2022

Natural Language Interfaces to Data

Recent advances in NLU and NLP have resulted in renewed interest in natural language interfaces to data, which provide an easy mechanism for non-technical users to access and query the data. While early systems evolved from keyword search and focused on simple factual queries, the complexity of both the input sentences as well as the generated SQL queries has evolved over time. More recently, there has also been a lot of focus on using conversational interfaces for data analytics, empowering a line of non-technical users with quick insights into the data. There are three main challenges in natural language querying (NLQ): (1) identifying the entities involved in the user utterance, (2) connecting the different entities in a meaningful way over the underlying data source to interpret user intents, and (3) generating a structured query in the form of SQL or SPARQL. There are two main approaches for interpreting a user's NLQ. Rule-based systems make use of semantic indices, ontologies, and KGs to identify the entities in the query, understand the intended relationships between those entities, and utilize grammars to generate the target queries. With the advances in deep learning (DL)-based language models, there have been many text-to-SQL approaches that try to interpret the query holistically using DL models. Hybrid approaches that utilize both rule-based techniques as well as DL models are also emerging by combining the strengths of both approaches. Conversational interfaces are the next natural step to one-shot NLQ by exploiting query context between multiple turns of conversation for disambiguation. In this article, we review the background technologies that are used in natural language interfaces, and survey the different approaches to NLQ. We also describe conversational interfaces for data analytics and discuss several benchmarks used for NLQ research and evaluation.

cs.DB↗

arXiv2022

Scaling Knowledge Graph Embedding Models

Developing scalable solutions for training Graph Neural Networks (GNNs) for link prediction tasks is challenging due to the high data dependencies which entail high computational cost and huge memory footprint. We propose a new method for scaling training of knowledge graph embedding models for link prediction to address these challenges. Towards this end, we propose the following algorithmic strategies: self-sufficient partitions, constraint-based negative sampling, and edge mini-batch training. Both, partitioning strategy and constraint-based negative sampling, avoid cross partition data transfer during training. In our experimental evaluation, we show that our scaling solution for GNN-based knowledge graph embedding models achieves a 16x speed up on benchmark datasets while maintaining a comparable model performance as non-distributed methods on standard metrics.

cs.LG↗

arXiv2021

BI-REC: Guided Data Analysis for Conversational Business Intelligence

Conversational interfaces to Business Intelligence (BI) applications enable data analysis using a natural language dialog in small incremental steps. To truly unleash the power of conversational BI to democratize access to data, a system needs to provide effective and continuous support for data analysis. In this paper, we propose BI-REC, a conversational recommendation system for BI applications to help users accomplish their data analysis tasks. We define the space of data analysis in terms of BI patterns, augmented with rich semantic information extracted from the OLAP cube definition, and use graph embeddings learned using GraphSAGE to create a compact representation of the analysis state. We propose a two-step approach to explore the search space for useful BI pattern recommendations. In the first step, we train a multi-class classifier using prior query logs to predict the next high-level actions in terms of a BI operation (e.g., {\em Drill-Down} or {\em Roll-up}) and a measure that the user is interested in. In the second step, the high-level actions are further refined into actual BI pattern recommendations using collaborative filtering. This two-step approach allows us to not only divide and conquer the huge search space, but also requires less training data. Our experimental evaluation shows that BI-REC achieves an accuracy of 83% for BI pattern recommendations and up to 2X speedup in latency of prediction compared to a state-of-the-art baseline. Our user study further shows that BI-REC provides recommendations with a precision@3 of 91.90% across several different analysis tasks.

cs.DB↗

arXiv2021

Medical Entity Disambiguation Using Graph Neural Networks

Medical knowledge bases (KBs), distilled from biomedical literature and regulatory actions, are expected to provide high-quality information to facilitate clinical decision making. Entity disambiguation (also referred to as entity linking) is considered as an essential task in unlocking the wealth of such medical KBs. However, existing medical entity disambiguation methods are not adequate due to word discrepancies between the entities in the KB and the text snippets in the source documents. Recently, graph neural networks (GNNs) have proven to be very effective and provide state-of-the-art results for many real-world applications with graph-structured data. In this paper, we introduce ED-GNN based on three representative GNNs (GraphSAGE, R-GCN, and MAGNN) for medical entity disambiguation. We develop two optimization techniques to fine-tune and improve ED-GNN. First, we introduce a novel strategy to represent entities that are mentioned in text snippets as a query graph. Second, we design an effective negative sampling strategy that identifies hard negative samples to improve the model's disambiguation capability. Compared to the best performing state-of-the-art solutions, our ED-GNN offers an average improvement of 7.3% in terms of F1 score on five real-world datasets.

cs.IR↗

arXiv2021

To Share, or not to Share Online Event Trend Aggregation Over Bursty Event Streams

Complex event processing (CEP) systems continuously evaluate large workloads of pattern queries under tight time constraints. Event trend aggregation queries with Kleene patterns are commonly used to retrieve summarized insights about the recent trends in event streams. State-of-art methods are limited either due to repetitive computations or unnecessary trend construction. Existing shared approaches are guided by statically selected and hence rigid sharing plans that are often sub-optimal under stream fluctuations. In this work, we propose a novel framework Hamlet that is the first to overcome these limitations. Hamlet introduces two key innovations. First, Hamlet adaptively decides whether to share or not to share computations depending on the current stream properties at run time to harvest the maximum sharing benefit. Second, Hamlet is equipped with a highly efficient shared trend aggregation strategy that avoids trend construction. Our experimental study on both real and synthetic data sets demonstrates that Hamlet consistently reduces query latency by up to five orders of magnitude compared to the state-of-the-art approaches.

cs.DB↗

arXiv2020

Event Trend Aggregation Under Rich Event Matching Semantics

Streaming applications from health care analytics to algorithmic trading deploy Kleene queries to detect and aggregate event trends. Rich event matching semantics determine how to compose events into trends. The expressive power of state-of-the-art systems remains limited in that they do not support the rich variety of these semantics. Worse yet, they suffer from long delays and high memory costs because they opt to maintain aggregates at a fine granularity. To overcome these limitations, our Coarse-Grained Event Trend Aggregation (Cogra) approach supports this rich diversity of event matching semantics within one system. Better yet, Cogra incrementally maintains aggregates at the coarsest granularity possible for each of these semantics. In this way, Cogra minimizes the number of aggregates -- reducing both time and space complexity. Our experiments demonstrate that Cogra achieves up to four orders of magnitude speed-up and up to eight orders of magnitude memory reduction compared to state-of-the-art approaches.

cs.DB↗

arXiv2020

GRETA: Graph-based Real-time Event Trend Aggregation

Streaming applications from algorithmic trading to traffic management deploy Kleene patterns to detect and aggregate arbitrarily-long event sequences, called event trends. State-of-the-art systems process such queries in two steps. Namely, they first construct all trends and then aggregate them. Due to the exponential costs of trend construction, this two-step approach suffers from both a long delays and high memory costs. To overcome these limitations, we propose the Graph-based Real-time Event Trend Aggregation (Greta) approach that dynamically computes event trend aggregation without first constructing these trends. We define the Greta graph to compactly encode all trends. Our Greta runtime incrementally maintains the graph, while dynamically propagating aggregates along its edges. Based on the graph, the final aggregate is incrementally updated and instantaneously returned at the end of each query window. Our Greta runtime represents a win-win solution, reducing both the time complexity from exponential to quadratic and the space complexity from exponential to linear in the number of events. Our experiments demonstrate that Greta achieves up to four orders of magnitude speed-up and up to 50--fold memory reduction compared to the state-of-the-art two-step approaches.

cs.DS↗

arXiv2020

Sharon: Shared Online Event Sequence Aggregation

Streaming systems evaluate massive workloads of event sequence aggregation queries. State-of-the-art approaches suffer from long delays caused by not sharing intermediate results of similar queries and by constructing event sequences prior to their aggregation. To overcome these limitations, our Shared Online Event Sequence Aggregation (Sharon) approach shares intermediate aggregates among multiple queries while avoiding the expensive construction of event sequences. Our Sharon optimizer faces two challenges. One, a sharing decision is not always beneficial. Two, a sharing decision may exclude other sharing opportunities. To guide our Sharon optimizer, we compactly encode sharing candidates, their benefits, and conflicts among candidates into the Sharon graph. Based on the graph, we map our problem of finding an optimal sharing plan to the Maximum Weight Independent Set (MWIS) problem. We then use the guaranteed weight of a greedy algorithm for the MWIS problem to prune the search of our sharing plan finder without sacrificing its optimality. The Sharon optimizer is shown to produce sharing plans that achieve up to an 18-fold speed-up compared to state-of-the-art approaches.

cs.DB↗