arXiv subjects

Whanhee Cho

Publications and source records attributed to Whanhee Cho.

3 recordsLinked to original sources

arXiv2026

Example-Driven Intent Synthesis for Constrained Data Bundle Retrieval: Focused Text Snippet Extraction and Beyond

Selecting a bundle of items that collectively satisfies constraints is a fundamental task across databases, recommender systems, and text summarization. Unlike traditional retrieval that returns individual or top-k items, bundle retrieval is inherently combinatorial and, in general, NP-hard. Although package queries can efficiently retrieve bundles given a well-formed query, two key user-centric challenges remain: (1) expressing and tuning multi-dimensional bundle intent through a user-friendly interface, and (2) ensuring feasibility when the query yields empty results. We introduce Ex2Bundle, an Example-driven Bundle retrieval framework that enables users to specify their intent through example bundles and automatically synthesizes package queries that capture the intent implicit in those example bundles via aggregate constraints. Ex2Bundle also addresses a challenge unique to bundle retrieval: when inferred aggregate constraints are infeasible over the target data, our data-aware constraint relaxation minimally adjusts the constraint bounds while preserving alignment with user intent. We instantiate a specific application of focused text snippet extraction by example to demonstrate the efficacy of the Ex2Bundle framework. Extensive experiments over real-world datasets and a user study demonstrate that Ex2Bundle improves usability and consistently returns intent-aligned bundles even under distributional shifts of the target database.

cs.DB↗

arXiv2025

Data-Semantics-Aware Recommendation of Diverse Pivot Tables

Data summarization is essential to discover insights from large datasets. In a spreadsheets, pivot tables offer a convenient way to summarize tabular data by computing aggregates over some attributes, grouped by others. However, identifying attribute combinations that will result in useful pivot tables remains a challenge, especially for high-dimensional datasets. We formalize the problem of automatically recommending insightful and interpretable pivot tables, eliminating the tedious manual process. A crucial aspect of recommending a set of pivot tables is to diversify them. Traditional works inadequately address the table-diversification problem, which leads us to consider the problem of pivot table diversification. We present SAGE, a data-semantics-aware system for recommending k-budgeted diverse pivot tables, overcoming the shortcomings of prior work for top-k recommendations that cause redundancy. SAGE ensures that each pivot table is insightful, interpretable, and adaptive to the user's actions and preferences, while also guaranteeing that the set of pivot tables are different from each other, offering a diverse recommendation. We make two key technical contributions: (1) a data-semantics-aware model to measure the utility of a single pivot table and the diversity of a set of pivot tables, and (2) a scalable greedy algorithm that can efficiently select a set of diverse pivot tables of high utility, by leveraging data semantics to significantly reduce the combinatorial search space. Our extensive experiments on three real-world datasets show that SAGE outperforms alternative approaches, and efficiently scales to accommodate high-dimensional datasets. Additionally, we present several case studies to highlight SAGE's qualitative effectiveness over commercial software and Large Language Models (LLMs).

cs.DB↗

arXiv2024

ACCIO: Table Understanding Enhanced via Contrastive Learning with Aggregations

The attention to table understanding using recent natural language models has been growing. However, most related works tend to focus on learning the structure of the table directly. Just as humans improve their understanding of sentences by comparing them, they can also enhance their understanding by comparing tables. With this idea, in this paper, we introduce ACCIO, tAble understanding enhanCed via Contrastive learnIng with aggregatiOns, a novel approach to enhancing table understanding by contrasting original tables with their pivot summaries through contrastive learning. ACCIO trains an encoder to bring these table pairs closer together. Through validation via column type annotation, ACCIO achieves competitive performance with a macro F1 score of 91.1 compared to state-of-the-art methods. This work represents the first attempt to utilize pairs of tables for table embedding, promising significant advancements in table comprehension. Our code is available at https://github.com/whnhch/ACCIO/.

cs.CL↗