Pith. sign in

REVIEW 4 major objections 6 minor 84 references

Towards Automated Cross-domain Exploratory Data Analysis through Large Language Models

T0 review · 4 major / 6 minor · reviewed 2026-08-11 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read TiInsight claims that LLM-generated hierarchical schema summaries, not fine-tuning, are enough to reach near-state-of-the-art text-to-SQL and cross-domain exploratory data analysis.

desk verdict Credible industrial EDA system with plausible benchmark numbers, but HDC's contribution is never isolated and the 'SOTA' claim overstates the paper's own tables. read the letter →

arxiv 2412.07214 v3 pith:HCPQWN2R submitted 2024-12-10 cs.DB cs.AI

Jun-Peng Zhu , Boyan Niu , Peng Cai , Zheming Ni , Jianwei Wan , Kai Xu , Jiajun Huang , Shengbo Ma
show 6 more authors
Bing Wang Xuan Zhou Guanglei Bao Donghui Zhang Liu Tang Qi Liu
This is my paper · ORCID
classification cs.DBcs.AI
keywords ExploratorydataanalysisText-to-SQLLargelanguagemodelsDatabaseschemasummarizationHierarchicalcontextSQLgenerationvisualizationCross-domaingeneralization
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

TiInsight is an end-to-end SQL-based exploratory data analysis system that claims to generalize across data domains without per-domain fine-tuning. The paper's central claim is that condensing a complex database schema into three levels of LLM-written summaries—columns, tables, and databases—plus vector-database retrieval, gives a text-to-SQL component that reaches 86.3% execution accuracy on Spider with GPT-4 and state-of-the-art results on Bird. The same pipeline is reported to answer real-world EDA questions with an 82.3% execution success rate during six months of public use. The paper argues this matters because existing EDA and text-to-SQL methods are benchmark-oriented, need labeled data for fine-tuning, and cannot handle wide tables with hundreds of columns or ambiguous user questions.

What carries the argument

The central object is the hierarchical data context (HDC): three nested LLM-generated summaries of a database, at column, table, and database level, stored in a vector database. Column summaries are produced in parallel groups with retrieved domain knowledge and sample rows; table summaries combine a map-reduce description of wide tables with two-stage relationship detection; database summaries are built from the top-N most-connected tables and their inferred entities. HDC carries the argument because it replaces the raw schema in prompts, letting TiSQL retrieve only relevant tables and columns, fit inside context windows, and transfer across domains without fine-tuning.

What would settle it

A schema with a deliberately misleading or abbreviated column (for example, dod standing for day-over-day) where the column summary omits the expansion; if TiSQL cannot produce the correct SQL even though the raw schema comment contains the meaning, the HDC faithfulness assumption is falsified.

Watch

Extended reading notes

Core claim

The paper's discovery is that an open-world EDA system can be built from four LLM-driven stages—hierarchical data context (HDC) generation, question clarification and decomposition, text-to-SQL (TiSQL), and rule-based visualization (TiChart)—where HDC is the load-bearing novelty. HDC uses LLMs to summarize each column, then each table, then the database as a whole, storing these summaries in a vector database. TiSQL retrieves relevant tables and columns from these summaries, builds few-shot chain-of-thought prompts, and applies a self-refinement chain using EXPLAIN and execution feedback. In the paper's evaluation, this design without fine-tuning reaches 86.3% execution accuracy on Spider with GPT-4, outperforms DAIL-SQL on Bird, and achieves 82.3% execution success on real-world tasks.

Load-bearing premise

The pipeline assumes that the LLM-generated summaries of columns, tables, and databases preserve all the information needed to answer later questions, so anything the summaries omit or mistranslate is unrecoverable by the SQL stage.

Editorial extensions

If this is right

  • A text-to-SQL component that never sees the raw schema can match or beat fine-tuned systems: TiSQL reaches 86.3% on Spider and state-of-the-art execution accuracy on Bird with GPT-4.
  • Cross-domain EDA becomes feasible without per-domain retraining: HDC summarization plus vector retrieval is domain-agnostic, and the system reports 82.3% execution success across finance, retail, and gaming tasks in public testing.
  • Ambiguous or multi-step user questions can be handled by a clarification-and-decomposition stage that turns them into explicit sub-questions for SQL generation.
  • Chart selection can be rule-based and LLM-checked, so end-to-end text-to-visualization requires no specialized visualization language or reinforcement learning.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • If HDC summaries are the reason for the gains, the same summarization-and-retrieval pattern could be lifted into any NL2SQL or data-agent pipeline, not just TiSQL; a direct ablation comparing HDC against raw-schema prompting on the same backbone would test this.
  • The 82.3% real-world execution success rate is not directly comparable to benchmark accuracy because it counts executability, not correctness against gold SQL; a deployment study with labeled answers would be the natural next measurement.
  • The map-reduce summarization scheme suggests a scaling rule of thumb for enterprise schemas: grouping of 40 columns for GPT-4 and 80 for other models, with retrieval over summaries, could be adopted as a standard preprocessing step for wide tables.
  • Since HDC generation is one-time per database, its latency cost can be amortized; the paper's latency numbers imply that the economic bottleneck of such systems is LLM token cost, not summarization time.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 6 minor

Summary. The paper presents TiInsight, an LLM-based end-to-end system for cross-domain exploratory data analysis. The system has four stages: hierarchical data context (HDC) generation, which uses LLMs to produce column, table, and database summaries; question clarification and decomposition; TiSQL, a zero-shot text-to-SQL method that performs coarse- and fine-grained schema filtering and then applies a self-refinement chain; and TiChart, a rule-based visualization recommender. The central empirical claims are that TiSQL achieves 86.3% execution accuracy on Spider with GPT-4, that it achieves state-of-the-art execution accuracy on Bird (60.98% test EX), that it shows an 82.3% execution success rate in six months of public testing at PingCAP, and that a user study finds TiInsight comparable to human experts. The authors also report open-sourcing the system's APIs and evaluation code.

Significance. If the central claim holds, this is a practically significant systems result: it suggests that LLM-generated hierarchical schema summaries can make a no-fine-tuning text-to-SQL pipeline competitive with specialized and fine-tuned systems across domains, in a deployed enterprise setting. The paper's strengths are its evaluation on external public benchmarks (Spider and Bird), its production deployment evidence, and its release of code and APIs, which makes independent verification feasible. However, the contribution attribution is not yet established: the paper never isolates HDC's effect, and the accuracy margins over strong baselines are small. Because the cross-domain generalization claim rests on HDC, the missing ablation and missing schema-linking recall analysis are load-bearing gaps rather than cosmetic issues.

major comments (4)
  1. [§6.2, Tables 1 and 3] The paper attributes TiSQL's benchmark results 'primarily to HDC' (p. 9, col. 1) and the abstract credits HDC for cross-domain generalization, but no experiment isolates HDC's contribution. There is no ablation comparing TiSQL with HDC against TiSQL with raw schemas (or with HDC components disabled), and no schema-linking recall statistic showing that gold tables and columns survive the coarse-grained and fine-grained filters of Algorithm 4. Given the small margins over strong baselines (86.3% vs. 86.2% for DAIL-SQL + GPT-4 on Spider; 60.98% vs. 60.37% for SFT CodeS-15B on Bird test EX; and R-VES below CodeS-15B), the observed accuracy could plausibly reflect GPT-4's zero-shot capability and the self-refinement chain rather than HDC. This is load-bearing for the central cross-domain generalization claim and should be addressed with an ablation and/or a recall analysis before the attribution is made.
  2. [§3.1.1, §3.1.2, and Algorithm 4] HDC is lossy by construction: column summaries are generated on groups of 40-80 columns with only three randomly sampled rows per column (§3.1.1), table relationships are identified from at most the top-20 similar tables (§3.1.2), and TiSQL subsequently filters to the top-N retrieved tables (Algorithm 4). If a key column or foreign-key relationship is omitted or distorted in these summaries, downstream retrieval and SQL generation cannot recover it. The paper reports no fidelity metric for the summaries and no recall of gold tables/columns/joins on Spider or Bird. Since the stated motivation is complex real-world schemas with thousands of columns, the paper should measure information loss (e.g., gold-table/column recall after HDC + TabColFilter) or at least compare against a raw-schema condition on schemas that fit in the context window.
  3. [§6.2, public testing paragraph] The 82.3% figure reported from six months of public testing is an 'execution success rate' only, i.e., the generated SQL executed without error. It is not a measure of whether the SQL answered the user's question, and the paragraph does not define what counts as an EDA task, how failures were classified, or what baseline was used. As presented, this number does not support the abstract's claim of success on 'real-world EDA tasks' beyond syntactic executability; it should be relabeled or supplemented with a correctness or user-confirmation measure.
  4. [§6.3, Figures 6 and 7] The user study has 20 participants, 60 minutes, three datasets, and no statistical significance testing; the text states TiInsight is 'comparable' to the human expert based on overlapping confidence intervals. With this sample size and no paired tests or effect sizes, the claim in the abstract of 'remarkable performance compared to human experts' is not supported. At minimum, report per-participant paired comparisons and a significance test (or explicitly recast the finding as a descriptive usability observation).
minor comments (6)
  1. [Abstract and Table 4] The 'state-of-the-art performance on the Bird dataset' claim should be qualified: on R-VES (Table 4), TiSQL + GPT-4 (56.06) is below SFT CodeS-15B (56.73), so SOTA holds only for EX, not for R-VES.
  2. [§6.5 and Figure 10] Cost reporting is internally inconsistent and hard to read: the text says GPT-4 costs 'about $300' while Figure 10 shows values of 260.0 and 280.0, and Table 5's pricing cell is garbled (e.g., '$1.25 / 1M cached$0.15 / 1M tokens').
  3. [§3.1] The statement that 'the prompt used in this section is not included in the paper' is a reproducibility gap for the core HDC component; because the code is open-sourced, please either include the prompt in an appendix or point to the exact file and version in the repository.
  4. [§3.1.1] The random three-row sampling per column is not described with a seed or stability check; a short sensitivity analysis would assure readers that summaries do not depend on the particular sampled rows.
  5. [Figure 9] Latency results are reported as averages without variance or a per-component breakdown, making it hard to assess stability under concurrency.
  6. [References] Several references reuse the ACM template defaults (e.g., 'Received 20 February 2007') and contain 'Retrived' typos; these should be cleaned before camera-ready.

Circularity Check

0 steps flagged · score 1.0 of 10

No derivation step reduces to its own input; benchmark scores are external measurements, so circularity is minimal (score 1).

full rationale

The central quantitative claims — 86.3% EX on Spider, 60.98% EX and 56.06 R-VES on Bird, and 82.3% execution success in production — are measured against external public benchmarks and a deployment, not derived from the paper's own assumptions. HDC, TiSQL, and TiChart are presented as a pipeline whose components consume the schema summaries the system itself generates, but the benchmark evaluation does not inject those summaries into the gold answer. The hyperparameters (40/80 columns per group, similar_count=20, top-N=20/30) are stated as fixed engineering choices rather than fitted to the reported accuracy. The only self-citation is the authors' prior Chat2Query system, used as a user-study baseline; it does not justify TiSQL's accuracy. The paper's own text flags two limitations that bear on support rather than circularity: the HDC prompts are omitted ('Due to space constraints, the prompt used in this section is not included in the paper'), and no objective EDA benchmark exists ('Currently, no open-source benchmark is available for the objective evaluation of exploratory data analysis systems'). Additionally, the claim that TiSQL's Spider result 'is primarily due to HDC' (§6.2) is not backed by an ablation or schema-linking recall analysis, so the attribution of the margin to HDC is under-supported. That missing ablation is a completeness/evidence gap, not a circular reduction: the reported accuracies would remain measurements even if HDC contributed nothing. No fitted parameter is renamed as a prediction, and no load-bearing argument reduces to a self-citation chain. Score 1 reflects the minor self-citation and unsupported causal attribution, with no significant circularity.

Assumptions & free parameters 5 free parameters · 4 assumptions · 0 invented entities

The central claims rest on empirical hyperparameters (group sizes, retrieval counts, N) and on qualitative assumptions about the reliability of LLM summaries and vector retrieval. No new physical or formal entities are introduced.

free parameters (5)
  • HDC column group size = 40 columns for GPT-4, 80 for other models
    Set empirically to fit context windows; affects summarization granularity and cost (§3.1.1).
  • similar_count for table relationship = 20
    Number of candidate tables retrieved in coarse-grained search for relationship inference (§3.1.2).
  • top_N for entity extraction = 20
    Number of high-relationship tables used to infer database entities (§3.1.3).
  • N for TiSQL coarse-grained table retrieval = 30
    Number of candidate tables retrieved from vector DB during schema filtering (§4).
  • Rows sampled per column = 3
    Sample rows included in column summaries to help the LLM understand value types (§3.1.1).
assumptions (4)
  • domain assumption LLM-generated schema summaries preserve enough information for correct SQL generation
    HDC replaces raw schema with LLM summaries; if these are lossy, retrieval and SQL generation degrade (§3.1).
  • domain assumption Vector similarity retrieval identifies relevant tables and columns for a given question
    Used in table relationship inference and TiSQL schema filtering; no evaluation of retrieval quality (§4, Algorithm 4).
  • domain assumption Execution feedback through EXPLAIN and runtime errors improves SQL correctness
    Self-refine chain assumes error feedback can be used by the LLM to fix SQL (§4).
  • domain assumption Rule-based chart selection with LLM verification matches user preferences
    TiChart rules derived from PingCAP user preferences and encoded in CoT prompts (§5).

how reviews work

0 comments
Cite this review

Pith. "Pith review of Towards Automated Cross-domain Exploratory Data Analysis through Large Language Models." pith.science (2026). https://pith.science/paper/HCPQWN2R

@misc{pith2026241207214,
  author       = {Pith},
  title        = {Pith review of: Towards Automated Cross-domain Exploratory Data Analysis through Large Language Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/HCPQWN2R}},
  note         = {Machine review of arXiv:2412.07214}
}
read the original abstract

Exploratory data analysis (EDA), coupled with SQL, is essential for data analysts involved in data exploration and analysis. However, data analysts often encounter two primary challenges: (1) the need to craft SQL queries skillfully, and (2) the requirement to generate suitable visualization types that enhance the interpretation of query results. Due to its significance, substantial research efforts have been made to explore different approaches to address these challenges, including leveraging large language models (LLMs). However, existing methods fail to meet real-world data exploration requirements primarily due to (1) complex database schema; (2) unclear user intent; (3) limited cross-domain generalization capability; and (4) insufficient end-to-end text-to-visualization capability. This paper presents TiInsight, an automated SQL-based cross-domain exploratory data analysis system. First, we propose hierarchical data context (i.e., HDC), which leverages LLMs to summarize the contexts related to the database schema, which is crucial for open-world EDA systems to generalize across data domains. Second, the EDA system is divided into four components (i.e., stages): HDC generation, question clarification and decomposition, text-to-SQL generation (i.e., TiSQL), and data visualization (i.e., TiChart). Finally, we implemented an end-to-end EDA system with a user-friendly GUI interface in the production environment at PingCAP. We have also open-sourced all APIs of TiInsight to facilitate research within the EDA community. Through extensive evaluations by a real-world user study, we demonstrate that TiInsight offers remarkable performance compared to human experts. Specifically, TiSQL achieves an execution accuracy of 86.3% on the Spider dataset using GPT-4. It also demonstrates state-of-the-art performance on the Bird dataset.

Figures

Figures reproduced from arXiv: 2412.07214 by the authors.

Figure 1
Figure 1. Overall architecture of TiInsight. TiInsight to facilitate large-scale and complex database schemas during the HDC generation. The generated data context is stored in vector databases. (3) Building on HDC, we propose a cross-domain text-to-SQL method named TiSQL. The TiSQL explores various prompting tech￾niques to enhance the accuracy of text-to-SQL generation. Building on this foundation, we introduce TiChart, whic… view at source ↗
Figure 2
Figure 2. Hierarchical data context generation of TiInsight. [PITH_FULL_IMAGE:figures/full_fig_p004_2.png] view at source ↗
Figure 3
Figure 3. An illustration of map-reduce for table description. [PITH_FULL_IMAGE:figures/full_fig_p005_3.png] view at source ↗
Figures from the paper (4 more)
Figure 4
Figure 4. Figure 4: The self-refine chain of TiSQL. clarified task. (2) Map keywords to columns and tables: Match the extracted keywords to specific columns and tables within the schemas, determining the appropriate column and table names of each keyword. (3) Generate SQL query: Using the…
Figure 5
Figure 5. Figure 5: Spider dev result analysis. method. This is primarily due to HDC, which aims to maximize the use of database schema information, thereby achieving full utiliza￾tion of schema information in TiSQL. This further demonstrates the strong contextual learning and summarizati…
Figure 6
Figure 6. Figure 6: Relevance scores of the user study. 0 1 2 3 4 5 Avg Completeness Score Human Expert TiInsight Chat2Query ChatGPT (a) Completeness 0 1 2 3 4 5 Avg Understandability Score Human Expert TiInsight Chat2Query ChatGPT (b) Understandability [PITH_FULL_IMAGE:figures/full_fig_…
Figure 8
Figure 8. Figure 8: The prompt for ChatGPT. ChatGPT exhibited the worst, as users could not execute SQL statements directly during data exploration, necessitating extra effort to verify the generated SQL. It cannot generate visualization charts and only provides text descriptions of the c…

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

84 extracted references · 56 canonical work pages

  1. [1]

    [n. d.]. Act on AI-powered insights in your flow of work built on the Salesforce Platform with Agentforce. Retrived in May, 2024 from https://www.tableau.com/

    2024

  2. [2]

    [n. d.]. BIRD-SQL: A Big Bench for Large-Scale Database Grounded Text-to-SQLs. Retrived in May, 2024 from https://bird-bench.github.io/

    2024

  3. [3]

    [n. d.]. Chroma - the open-source embedding database. Retrived in May, 2024 from https://github.com/chroma-core/chroma

    2024

  4. [4]

    [n. d.]. Crude Oil WTI Futures. Retrived in May, 2024 from https://www.investing. com/commodities/crude-oil-historical-data

    2024

  5. [5]

    [n. d.]. Dow Jones Industrial Average. Retrived in May, 2024 from https://fred. stlouisfed.org/series/DJIA

    work page 2024

  6. [6]

    [n. d.]. Federal Funds Effective Rate. Retrived in May, 2024 from https://fred. stlouisfed.org/series/FEDFUNDS

    work page 2024

  7. [7]

    [n. d.]. Gold futures. Retrived in May, 2024 from https://www.investing.com/ commodities/gold-historical-data

    work page 2024

  8. [8]

    [n. d.]. GPT-4o mini: advancing cost-efficient intelligence. Retrived in May, 2024 from https://openai.com/index/gpt-4o-mini-advancing-cost-efficient- intelligence/

    work page 2024

Show all 84 references
  1. [9]

    [n. d.]. Hello GPT-4o. Retrived in May, 2024 from https://openai.com/index/hello- gpt-4o/

    2024

  2. [10]

    [n. d.]. The JavaScript library for bespoke data visualization. Retrived in May, 2024 from https://d3js.org/

    2024

  3. [11]

    [n. d.]. NASDAQ Composite Index. Retrived in May, 2024 from https://fred. stlouisfed.org/series/NASDAQCOM

    2024

  4. [12]

    [n. d.]. Pinecone serverless lets you deliver remarkable GenAI applications faster. Retrived in May, 2024 from https://www.pinecone.io/

    2024

  5. [13]

    [n. d.]. Power BI: Uncover powerful insights and turn them into impact. Retrived in May, 2024 from https://www.microsoft.com/en-us/power-platform/products/ power-bi

    2024

  6. [14]

    [n. d.]. Shanghai Shenzhen CSI 300. Retrived in May, 2024 from https://www. investing.com/indices/csi300

    2024

  7. [15]

    [n. d.]. Spider: Yale Semantic Parsing and Text-to-SQL Challenge. Retrived in May, 2024 from https://yale-lily.github.io/spider

    2024

  8. [16]

    [n. d.]. Unemployment Rate. Retrived in May, 2024 from https://fred.stlouisfed. org/series/UNRATE

    2024

  9. [17]

    [n. d.]. Vector Search (Beta) Overview. Retrived in May, 2024 from https://docs. pingcap.com/tidbcloud/vector-search-overview

    2024

  10. [18]

    Sihem Amer-Yahia. 2024. Intelligent Agents for Data Exploration. Proceedings of the VLDB Endowment 17, 12 (2024), 4521–4530

    2024

  11. [19]

    AI Anthropic. 2024. The claude 3 model family: Opus, sonnet, haiku. Claude-3 Model Card 1 (2024)

    2024

  12. [20]

    Christopher Baik, Zhongjun Jin, Michael Cafarella, and HV Jagadish. 2020. Duo- quest: A dual-specification system for expressive SQL queries. In Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data . 2319– 2329

    2020

  13. [21]

    Ben Bogin, Jonathan Berant, and Matt Gardner. 2019. Representing Schema Structure with Graph Neural Networks for Text-to-SQL Parsing. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics . 4560– 4565

    2019

  14. [22]

    Zhi Chen, Lu Chen, Yanbin Zhao, Ruisheng Cao, Zihan Xu, Su Zhu, and Kai Yu

  15. [23]

    DongHyun Choi, Myeong Cheol Shin, EungGyun Kim, and Dong Ryeol Shin. 2021. Ryansql: Recursively applying sketch-based slot fillings for complex text-to-sql in cross-domain databases. Computational Linguistics 47, 2 (2021), 309–332

    2021

  16. [24]

    Daniel Deutch and Amir Gilad. 2016. QPlain: Query by explanation. In 2016 IEEE 32nd International Conference on Data Engineering (ICDE) . IEEE, 1358–1361

    2016

  17. [25]

    Daniel Deutch, Amir Gilad, Tova Milo, Amit Mualem, and Amit Somech. 2022. FEDEX: An Explainability Framework for Data Exploration Steps. Proceedings of the VLDB Endowment 15, 13 (2022), 3854–3868

    2022

  18. [26]

    Jacob Devlin. 2018. Bert: Pre-training of deep bidirectional transformers for language understanding. arXiv preprint arXiv:1810.04805 (2018)

    2018 arXiv

  19. [27]

    Rui Ding, Shi Han, Yong Xu, Haidong Zhang, and Dongmei Zhang. 2019. Quickin- sights: Quick and automatic discovery of insights from multi-dimensional data. In Proceedings of the 2019 international conference on management of data . 317–332

    2019

  20. [28]

    Xuemei Dong, Chao Zhang, Yuhang Ge, Yuren Mao, Yunjun Gao, Jinshu Lin, Dongfang Lou, et al. 2023. C3: Zero-shot text-to-sql with chatgpt. arXiv preprint arXiv:2307.07306 (2023)

    2023 arXiv

  21. [29]

    Ori Bar El, Tova Milo, and Amit Somech. 2020. Towards Autonomous, Hands-Free Data Exploration.. In CIDR

    2020

  22. [30]

    Ju Fan, Zihui Gu, Songyue Zhang, Yuxin Zhang, Zui Chen, Lei Cao, Guoliang Li, Samuel Madden, Xiaoyong Du, and Nan Tang. 2024. Combining Small Language Models and Large Language Models for Zero-Shot NL2SQL. Proceedings of the VLDB Endowment 17, 11 (2024), 2750–2763

    2024

  23. [31]

    Anna Fariha and Alexandra Meliou. 2019. Example-driven query intent discov- ery: abductive reasoning using semantic similarity. Proceedings of the VLDB Endowment 12, 11 (2019), 1262–1275

    2019

  24. [32]

    Venkatesh Emani, Vivek Pandit, Victor Shnayder, Wenjing Wang, and Carlo Curino

    Avrilia Floratou, Fotis Psallidas, Fuheng Zhao, Shaleen Deep, Gunther Hagleither, Wangda Tan, Joyce Cahoon, Rana Alotaibi, Jordan Henkel, Abhik Singla, Alex Van Grootel, Brandon Chow, Kai Deng, Katherine Lin, Marcos Campos, K. Venkatesh Emani, Vivek Pandit, Victor Shnayder, We...

    2024

  25. [33]

    Dawei Gao, Haibin Wang, Yaliang Li, Xiuyu Sun, Yichen Qian, Bolin Ding, and Jin- gren Zhou. 2023. Text-to-sql empowered by large language models: A benchmark evaluation. arXiv preprint arXiv:2308.15363 (2023)

    2023 arXiv

  26. [34]

    Zihui Gu, Ju Fan, Nan Tang, Lei Cao, Bowen Jia, Sam Madden, and Xiaoyong Du

  27. [35]

    Jiaqi Guo, Zecheng Zhan, Yan Gao, Yan Xiao, Jian-Guang Lou, Ting Liu, and Dongmei Zhang. 2019. Towards Complex Text-to-SQL in Cross-Domain Database with Intermediate Representation. In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics . As...

    2019

  28. [36]

    Xinyi He, Mengyu Zhou, Xinrun Xu, Xiaojun Ma, Rui Ding, Lun Du, Yan Gao, Ran Jia, Xu Chen, Shi Han, et al. 2024. Text2analysis: A benchmark of table question answering with advanced data analysis and unclear queries. In Proceedings of the AAAI Conference on Artificial Intellig...

    2024

  29. [37]

    Stratos Idreos, Olga Papaemmanouil, and Surajit Chaudhuri. 2015. Overview of data exploration techniques. InProceedings of the 2015 ACM SIGMOD international conference on management of data . 277–281

    2015

  30. [38]

    George Katsogiannis-Meimarakis and Georgia Koutrika. 2023. A survey on deep learning approaches for text-to-SQL. The VLDB Journal (2023), 1–32

    2023

  31. [39]

    Hyeonji Kim, Byeong-Hoon So, Wook-Shin Han, and Hongrae Lee. 2020. Natural language to SQL: Where are we today? Proceedings of the VLDB Endowment 13, 10 (2020), 1737–1750

    2020

  32. [40]

    Tim Kraska. 2018. Northstar: An Interactive Data Science System. Proceedings of the VLDB Endowment 11, 12 (2018)

    2018

  33. [41]

    Doris Jung-Lin Lee, Dixin Tang, Kunal Agarwal, Thyne Boonmark, Caitlyn Chen, Jake Kang, Ujjaini Mukhopadhyay, Jerry Song, Micah Yong, Marti A Hearst, et al

  34. [42]

    Haoyang Li, Jing Zhang, Cuiping Li, and Hong Chen. 2023. Resdsql: Decoupling schema linking and skeleton parsing for text-to-sql. In Proceedings of the AAAI Conference on Artificial Intelligence, Vol. 37. 13067–13075

    2023

  35. [43]

    Haoyang Li, Jing Zhang, Hanbing Liu, Ju Fan, Xiaokang Zhang, Jun Zhu, Renjie Wei, Hongyan Pan, Cuiping Li, and Hong Chen. 2024. Codes: Towards build- ing open-source language models for text-to-sql. Proceedings of the ACM on Management of Data 2, 3 (2024), 1–28

    2024

  36. [44]

    Proceedings of the VLDB Endowment 15, 3 (2021), 727–738

    Lux: always-on visualization recommendations for exploratory dataframe workflows. Proceedings of the VLDB Endowment 15, 3 (2021), 727–738

    2021

  37. [45]

    Jinyang Li, Binyuan Hui, Ge Qu, Jiaxi Yang, Binhua Li, Bowen Li, Bailin Wang, Bowen Qin, Ruiying Geng, Nan Huo, et al . 2024. Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls. Advances in Neural Information Processing Sy...

    2024

  38. [46]

    Yuchen Li, Ju Fan, Yanhao Wang, and Kian-Lee Tan. 2018. Influence maximization on social graphs: A survey. IEEE Transactions on Knowledge and Data Engineering 30, 10 (2018), 1852–1872

    2018

  39. [47]

    Jinyang Li, Binyuan Hui, Reynold Cheng, Bowen Qin, Chenhao Ma, Nan Huo, Fei Huang, Wenyu Du, Luo Si, and Yongbin Li. 2023. Graphix-t5: Mixing pre-trained transformers with graph-aware layers for text-to-sql parsing. In Proceedings of the AAAI Conference on Artificial Intellige...

    2023

  40. [48]

    Yuyu Luo, Xuedi Qin, Nan Tang, and Guoliang Li. 2018. Deepeye: Towards automatic data visualization. In 2018 IEEE 34th international conference on data engineering (ICDE). IEEE, 101–112

    2018

  41. [49]

    Pingchuan Ma, Rui Ding, Shi Han, and Dongmei Zhang. 2021. Metainsight: Automatic discovery of structured knowledge for exploratory data analysis. In Proceedings of the 2021 international conference on management of data . 1262– 1274

    2021

  42. [50]

    Xi Victoria Lin, Richard Socher, and Caiming Xiong. 2020. Bridging Textual and Tabular Data for Cross-Domain Text-to-SQL Semantic Parsing. In Findings of the Association for Computational Linguistics: EMNLP 2020 . 4870–4888

    2020

  43. [51]

    Pingchuan Ma, Rui Ding, Shuai Wang, Shi Han, and Dongmei Zhang. 2023. XInsight: eXplainable Data Analysis Through The Lens of Causality. SIGMOD (2023), 1–27

    2023

  44. [52]

    Tova Milo and Amit Somech. 2018. Deep reinforcement-learning framework for exploratory data analysis. In Proceedings of the first international workshop on exploiting artificial intelligence techniques for data management . 1–4

    2018

  45. [53]

    Pingchuan Ma, Rui Ding, Shuai Wang, Shi Han, and Dongmei Zhang. 2023. InsightPilot: An LLM-empowered automated data exploration system. InProceed- ings of the 2023 Conference on Empirical Methods in Natural Language Processing: System Demonstrations. 346–352

    2023

  46. [54]

    OpenAI. [n. d.]. GPT-4. Retrived in May, 2024 from https://openai.com/research/ gpt-4. 13 Conference acronym ’XX, June 03–05, 2018, Woodstock, NY Jun-Peng Zhu et al

    2024

  47. [55]

    OpenAI. 2022. Introducing ChatGPT. Retrived in May, 2024 from https://openai. com/blog/chatgpt,

    2022

  48. [56]

    Tova Milo and Amit Somech. 2020. Automating exploratory data analysis via machine learning: An overview. In SIGMOD. 2617–2622

    2020

  49. [57]

    Mohammadreza Pourreza and Davood Rafiei. 2024. Din-sql: Decomposed in- context learning of text-to-sql with self-correction. Advances in Neural Informa- tion Processing Systems 36 (2024)

    2024

  50. [58]

    Mohammadreza Pourreza and Davood Rafiei. 2024. DTS-SQL: Decomposed Text-to-SQL with Small Large Language Models. arXiv preprint arXiv:2402.01117 (2024)

    2024 arXiv

  51. [59]

    Jinglin Peng, Weiyuan Wu, Brandon Lockhart, Song Bian, Jing Nathan Yan, Ling- hao Xu, Zhixuan Chi, Jeffrey M Rzeszotarski, and Jiannan Wang. 2021. Dataprep. eda: Task-centric exploratory data analysis for statistical modeling in python. In Proceedings of the 2021 International...

    2021

  52. [60]

    Ge Qu, Jinyang Li, Bowen Li, Bowen Qin, Nan Huo, Chenhao Ma, and Reynold Cheng. 2024. Before Generation, Align it! A Novel and Effective Strategy for Miti- gating Hallucinations in Text-to-SQL Generation. arXiv preprint arXiv:2405.15307 (2024)

    2024 arXiv

  53. [61]

    Diptikalyan Saha, Avrilia Floratou, Karthik Sankaranarayanan, Umar Farooq Minhas, Ashish R Mittal, and Fatma Özcan. 2016. ATHENA: an ontology-driven system for natural language querying over relational data stores. Proceedings of the VLDB Endowment 9, 12 (2016), 1209–1220

    2016

  54. [62]

    Xuedi Qin, Yuyu Luo, Nan Tang, and Guoliang Li. 2020. Making data visualization more efficient and effective: a survey. The VLDB Journal 29, 1 (2020), 93–117

    2020

  55. [63]

    Jaydeep Sen, Chuan Lei, Abdul Quamar, Fatma Özcan, Vasilis Efthymiou, Ayushi Dalmia, Greg Stager, Ashish Mittal, Diptikalyan Saha, and Karthik Sankara- narayanan. 2020. Athena++ natural language querying for complex nested sql queries. Proceedings of the VLDB Endowment 13, 12 ...

    2020

  56. [64]

    Leixian Shen, Enya Shen, Yuyu Luo, Xiaocong Yang, Xuming Hu, Xiongshuai Zhang, Zhiwei Tai, and Jianmin Wang. 2022. Towards natural language interfaces for data visualization: A survey. IEEE transactions on visualization and computer graphics 29, 6 (2022), 3121–3144

    2022

  57. [65]

    Torsten Scholak, Nathan Schucher, and Dzmitry Bahdanau. 2021. PICARD: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models. In Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing. 9895–9901

    2021

  58. [66]

    Bo Tang, Shi Han, Man Lung Yiu, Rui Ding, and Dongmei Zhang. 2017. Extracting top-k insights from multi-dimensional data. In Proceedings of the 2017 ACM international conference on management of data . 1509–1524

    2017

  59. [67]

    Jiawei Tang, Yuyu Luo, Mourad Ouzzani, Guoliang Li, and Hongyang Chen. 2022. Sevi: Speech-to-visualization through neural machine translation. In Proceedings of the 2022 International Conference on Management of Data . 2353–2356

    2022

  60. [68]

    Chris Stolte, Diane Tang, and Pat Hanrahan. 2002. Polaris: A system for query, analysis, and visualization of multidimensional relational databases. IEEE Trans- actions on Visualization and Computer Graphics 8, 1 (2002), 52–65

    2002

  61. [69]

    Manasi Vartak, Sajjadur Rahman, Samuel Madden, Aditya Parameswaran, and Neoklis Polyzotis. 2015. Seedb: Efficient data-driven visualization recommenda- tions to support visual analytics. In Proceedings of the VLDB Endowment Interna- tional Conference on Very Large Data Bases ,...

    2015

  62. [70]

    Bing Wang, Changyu Ren, Jian Yang, Xinnian Liang, Jiaqi Bai, Linzheng Chai, Zhao Yan, Qian-Wen Zhang, Di Yin, Xing Sun, and Zhoujun Li. 2024. MAC-SQL: A Multi-Agent Collaborative Framework for Text-to-SQL. arXiv:2312.11242 [cs.CL]

    2024 arXiv

  63. [71]

    Hugo Touvron, Thibaut Lavril, Gautier Izacard, Xavier Martinet, Marie-Anne Lachaux, Timothée Lacroix, Baptiste Rozière, Naman Goyal, Eric Hambro, Faisal Azhar, et al. 2023. Llama: Open and efficient foundation language models. arXiv preprint arXiv:2302.13971 (2023)

    2023 arXiv

  64. [72]

    Aoyu Wu, Yun Wang, Xinhuan Shu, Dominik Moritz, Weiwei Cui, Haidong Zhang, Dongmei Zhang, and Huamin Qu. 2021. Ai4vis: Survey on artificial intelligence approaches for data visualization. IEEE Transactions on Visualization and Computer Graphics 28, 12 (2021), 5049–5070

    2021

  65. [73]

    Chunyang Xiao, Marc Dymetman, and Claire Gardent. 2016. Sequence-based structured prediction for semantic parsing. In Annual meeting of the Association for Computational Linguistics (ACL). 1341–1350

    2016

  66. [74]

    Kanit Wongsuphasawat, Zening Qu, Dominik Moritz, Riley Chang, Felix Ouk, Anushka Anand, Jock Mackinlay, Bill Howe, and Jeffrey Heer. 2017. Voyager 2: Augmenting visual analysis with partial view specifications. In Proceedings of the 2017 chi conference on human factors in comp...

    2017

  67. [75]

    Xiaojun Xu, Chang Liu, and Dawn Song. 2017. Sqlnet: Generating structured queries from natural language without reinforcement learning. arXiv preprint arXiv:1711.04436 (2017)

    2017 arXiv

  68. [76]

    Tao Yu, Rui Zhang, Kai Yang, Michihiro Yasunaga, Dongxu Wang, Zifan Li, James Ma, Irene Li, Qingning Yao, Shanelle Roman, et al. 2018. Spider: A Large-Scale Human-Labeled Dataset for Complex and Cross-Domain Semantic Parsing and Text-to-SQL Task. In Proceedings of the 2018 Con...

    2018

  69. [77]

    Yupeng Xie, Yuyu Luo, Guoliang Li, and Nan Tang. 2024. HAIChart: Human and AI Paired Visualization System. arXiv preprint arXiv:2406.11033 (2024)

    2024 arXiv

  70. [78]

    Yunjia Zhang, Jordan Henkel, Avrilia Floratou, Joyce Cahoon, Shaleen Deep, and Jignesh M Patel. 2023. ReAcTable: Enhancing ReAct for Table Question Answering. arXiv preprint arXiv:2310.00815 (2023)

    2023 arXiv

  71. [79]

    Victor Zhong, Caiming Xiong, and Richard Socher. 2017. Seq2sql: Generating structured queries from natural language using reinforcement learning. arXiv preprint arXiv:1709.00103 (2017)

    2017 arXiv

  72. [80]

    Chao Zhang, Yuren Mao, Yijiang Fan, Yu Mi, Yunjun Gao, Lu Chen, Dongfang Lou, and Jinshu Lin. 2024. FinSQL: Model-Agnostic LLMs-based Text-to-SQL Framework for Financial Analysis. arXiv preprint arXiv:2401.10506 (2024)

    2024 arXiv

  73. [81]

    Jun-Peng Zhu, Peng Cai, Boyan Niu, Zheming Ni, Kai Xu, Jiajun Huang, Jianwei Wan, Shengbo Ma, Bing Wang, Donghui Zhang, Liu Tang, and Qi Liu. 2024. Chat2Query: A Zero-Shot Automatic Exploratory Data Analysis System with Large Language Models. In 2024 IEEE 40th International Co...

    2024

  74. [83]

    Mengyu Zhou, Qingtao Li, Xinyi He, Yuejiang Li, Yibo Liu, Wei Ji, Shi Han, Yining Chen, Daxin Jiang, and Dongmei Zhang. 2021. Table2Charts: recommending charts by learning shared table representations. In Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & D...

    2021

  75. [2021]

    In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies

    ShadowGNN: Graph Projection Neural Network for Text-to-SQL Parser. In Proceedings of the 2021 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies . 5567–5577

    2021

  76. [2023]

    SIGMOD (2023), 1–28

    Few-shot Text-to-SQL Translation using Structure and Content Prompt Learning. SIGMOD (2023), 1–28

    2023

Pith tools

Reviewed August 11, 2026 · model on record in the stance chip above.