Pith. sign in

REVIEW 4 major objections 7 minor 188 references

Revisiting Data Analysis with Pre-trained Foundation Models

T0 review · 4 major / 7 minor · reviewed 2026-08-10 · deepseek-v4-flash

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

Pith's one-line read This paper argues that pre-trained foundation models can systematically strengthen the utility and scalability of data analysis by enhancing reasoning, accessibility, data quality, and automation at every stage.

desk verdict A broad, useful survey of foundation models in data analysis whose PAC-learning framing is more rhetorical than substantive; the taxonomy stands on its own and the paper deserves revision, not rejection. read the letter →

arxiv 2501.01631 v1 pith:2VUQAIRS submitted 2025-01-03 cs.DB

Chen Liang , Donghua Yang , Zheng Liang , Zhiyu Liang , Tianle Zhang , Boyu Xiao , Yuqing Yang , Wenqi Wang
show 1 more author
Hongzhi Wang
This is my paper · ORCID
classification cs.DB
keywords dataanalysispre-trainedfoundationmodelsreasoningqualityaccessibilityautomationAutoMLdomain-specificlanguages
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

The paper's thesis is that pre-trained foundation models (PFMs) are not just another tool for data analysis but a way to systematically rewrite the entire pipeline: data preparation, exploratory analysis, implementation, and assessment. It organizes the field into four optimization levers — scalable reasoning, accessibility, data-quality management, and automation — and shows how each lever is already being pulled by existing PFM-based methods. The payoff, if the thesis holds, is that ad-hoc, hand-built analysis solutions can be replaced by systems that understand natural-language goals, generate and repair code, clean and synthesize data, and explain their own output. The paper also lists its own soft spots in Section 4 — inference cost, transferability, consistency, and trust — as open problems rather than refutations.

What carries the argument

The load-bearing structure is a cross-product between the four tasks of data analysis — data preparation, exploratory data analysis, implementation, and assessment — and four optimization dimensions that PFMs supply: scalable reasoning, accessibility, interpretability and editability, data quality, and automation (Sections 3.1–3.4). Reasoning is split into induction (from cases to rules) and deduction (from rules to cases), with consolidation of domain-specific languages (DSLs) as the bridge that makes partial solutions reusable. The conceptual lens is PAC learning, with its concept class $C$, hypothesis space $H$, distribution $D$, and learning algorithm $L$; the paper reads in-context learning as composing approximately correct hypotheses and uses this lens to sort methods by which component of the learning problem they improve.

What would settle it

A controlled benchmark would settle it: take a data-analysis workflow with increasing nesting depth (schema match, then join, then aggregation, then formal verification of the result), run it with state-of-the-art PFMs, and measure success rate and calibration as depth grows; if PFMs perform no better than a random baseline once the nesting depth exceeds two or three while still costing orders of magnitude more compute, the central claim collapses.

Watch

Extended reading notes

Core claim

The paper claims that PFMs introduce multi-dimensional advantages into data analysis because they can name and manipulate symbolic and formal concepts, compose probably-approximately-correct (PAC) hypotheses during in-context learning, and generalize across modalities. It argues these abilities let PFMs 'commute' specific tasks with the intelligent essence of earlier tools and theories: a query in natural language becomes executable SQL, a messy table becomes a cleaned and joined dataset, a code repository becomes a reusable library, and a statistical conclusion becomes a formally checked statement. The claimed result is a systematic optimization of utility and scalability across four task groups, with Section 3 mapping representative methods onto four optimization dimensions: scalable reasoning, accessibility of data and models, data-quality optimization, and automation (AutoML).

Load-bearing premise

The whole value proposition rests on the premise that PFMs can genuinely understand and combine concepts built from nested, repeated structures, and that their in-context learning yields approximately correct hypotheses; if that fails, the promised optimizations lose their foundation.

Editorial extensions

If this is right

  • Data analysis interfaces will shift from imperative programming to natural-language-declared goals, with PFMs routing to relevant data and methods and generating executable artifacts like SQL, code, and formal proofs.
  • Data preparation and cleaning become iterative, agent-driven processes — covering cleaning, schema matching, and joinability discovery — rather than one-shot manual wrangling.
  • AutoML expands beyond feature engineering and hyperparameter search into algorithm selection, code evolution, and domain-knowledge injection via PFMs.
  • Assessment of analyzed results can incorporate formal verification and satisfiability-modulo-theories checks, making conclusions about consistency and completeness more rigorous.
  • Maintaining and scaling analysis systems improves because consolidated DSLs and documented code repositories make partial solutions reusable across problems.

Reading between the lines

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

  • The paper's 'commuting' claim implies a testable prediction: for any well-specified data-analysis task, a PFM pipeline with in-context demonstrations should match or beat a hand-tuned imperative pipeline on accuracy, not just on development cost; the paper does not run such a head-to-head benchmark.
  • The four optimization dimensions are probably not independent — for example, automating cleaning (data quality) also tightens induction, which is the premise of the PAC framing — so a combined benchmark could measure whether improvements compound or saturate.
  • The practical payoff hinges on cost: the paper's own Section 4 cites daily operating costs in the hundreds of thousands of dollars for large models, so the systematic-optimization vision would need those costs to fall by orders of magnitude before it is deployable for most organizations.
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 / 7 minor

Summary. This paper is a survey that revisits the data analysis pipeline—data preparation, exploratory data analysis, implementation of models and methods, and assessment—and argues that pre-trained foundation models (PFMs) can systematically optimize every stage through four broad dimensions: reasoning, accessibility, data quality, and automation. The paper organizes a large set of recent references into an optimization framework, introduces PAC-learning notions in Sections 3.2 and 3.4 as a possible theoretical lens, and concludes with challenges and future directions. The central claim is that PFMs 'systematically strengthen the utility and scalability of data analysis' and can 'commute specific tasks with the essence of intelligent ability provided by previous tools and theories.'

Significance. If the central claim is accepted, the paper provides a useful organizing framework and a broad, current map of PFM-for-data-analysis research, with many recent VLDB, ICML, and NeurIPS references. A clear strength is that the paper does not ignore limitations: Section 4 explicitly discusses inference cost, domain generalization, consistency problems, and trust issues. As a survey, however, its value depends on accurately representing the cited works and on internally consistent reasoning. The paper's own Section 4 acknowledges failure modes that are in tension with the strong capability claims in Section 1.2, so the significance of the roadmap is currently undercut by overclaiming that goes beyond the cited evidence.

major comments (4)
  1. [Section 1.2] The claim that PFMs are 'composing probably approximately correct (PAC) hypothesis during in-context learning' is load-bearing for the paper's thesis that PFMs 'systematically' optimize data analysis, but it is asserted without proof or precise citation. The one provable in-context-learning result cited later ([7]) covers narrow linear settings, not the nested/recursive compositional reasoning invoked in the same section. Please qualify this claim to distinguish empirically observed behavior from theoretically proven guarantees, and cite the scope of [7] accurately.
  2. [Sections 3.2.2 and 3.4] Eq. (1) and the efficient-PAC definition in Section 3.4 are never connected to any PFM behavior. No theorem, bound, or empirical result shows that a PFM satisfies Eq. (1) for any data-analysis concept class, nor that any concept class is efficiently PAC learnable by a PFM in the sense of Section 3.4. The PAC language is therefore rhetorical rather than substantive, and since the 'systematic optimization' argument rests on this framing, please either provide a concrete statement of the intended PAC-style guarantee with appropriate caveats, or explicitly reframe the discussion as an analogy and state what would need to be proven.
  3. [Section 4] The limitations conceded in Section 4—hallucinations, prompt-order sensitivity, and susceptibility to irrelevant context—are in direct tension with the Section 1.2 claim that PFMs compose PAC hypotheses during in-context learning. Under a PAC-style guarantee with error budget epsilon and confidence delta, these failure modes would violate the guarantee. The paper should explain under what conditions (for example, with verification, in restricted domains, or with measured error rates) the PAC-style claim is intended to hold, or it should retract the claim.
  4. [Section 1.2] The statement that PFMs 'understand and compose complex semantics and concepts defined by nested or recursively built syntax structures' is supported by a citation to [3], which studies context-free grammar in a synthetic setup, and the follow-up claim that PFMs 'can approximately correctly choose operations and morphisms to build complex concepts' cites [16], a general few-shot-learning paper. Neither citation establishes the capability at the level claimed. Please tighten the claims to match the cited evidence or add citations that directly support the specific capability.
minor comments (7)
  1. [Figure 1 caption] The caption contains a typo: 'PMF-enhanced reasoning' should be 'PFM-enhanced reasoning.'
  2. [Sections 1.1 and 2.1, and Figure 2 caption] The terminology is inconsistent: the text sometimes refers to 'Data Preparation' and sometimes to 'Data Management' for the same task. Please align these labels.
  3. [Abstract] The phrase 'grounded with a large amount of grounded data' is redundant and ambiguous; please rephrase.
  4. [Section 2.1] There is a typo: 'Benefiting from FPMs' should be 'Benefiting from PFMs.'
  5. [Section 3.3.2] The sentence 'PFMs play a significant role in data cleaning by detecting and correcting errors in datasets ... [88]' cites a data-wrangling code-generation paper; either describe what [88] actually contributes or cite a more directly relevant evaluation of error detection.
  6. [References] References [73] and [74] are duplicate entries for the same work, 'PAC Prediction Sets for Large Language Models of Code.' Please merge them.
  7. [Section 4] The claim that costs 'make LLM technology unattainable for most organizations' is overstated in light of cheaper models and APIs; please soften the wording to match the cited cost estimates.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is a literature survey whose claims are supported by external citations and qualitative synthesis, not by fitting or self-referential derivation.

full rationale

The manuscript is a systematic review of pre-trained foundation models for data analysis. It does not derive a mathematical result, fit parameters, or produce a testable prediction from its own inputs, so the standard circularity patterns (self-definitional definitions, fitted inputs called predictions, load-bearing self-citation chains, ansatz smuggled via citation, or renaming a known result) do not apply. The paper's central claims—for example that PFMs can 'understand and compose complex semantics and concepts defined by nested or recursively built syntax structures' (Section 1.2) and can 'compose probably approximately correct (PAC) hypothesis during in-context learning' (Section 1.2)—are asserted as premises or as summaries of external literature, not derived from the paper's own framework. The PAC-learning language in Sections 3.2 and 3.4 is used as an organizing vocabulary rather than as a proof that PFMs satisfy PAC guarantees; no equation in the paper equates a PFM output to a fitted quantity by construction. The paper also candidly lists limitations in Section 4, including hallucination and prompt sensitivity, which further indicates that the authors are not presenting a forced or definitionally guaranteed optimization result. The survey does not rely on self-citations: no load-bearing argument is justified exclusively by the present authors' prior work. Therefore, the appropriate finding is no significant circularity, with a score of 0.

Assumptions & free parameters 0 free parameters · 3 assumptions · 0 invented entities

The paper introduces no new entities, parameters, or measured quantities. It is a review that relies on prior literature and on its own classification scheme. The assumptions listed are the structural choices that shape the survey's presentation.

assumptions (3)
  • domain assumption The four-way decomposition of data analysis tasks (preparation, exploration, implementation, assessment) is complete and adequate.
    The entire organization of the survey rests on this taxonomy, introduced in Section 1.1 and used to structure the paper. The paper asserts this decomposition without justifying why these four categories are exhaustive.
  • ad hoc to paper PFMs possess the capabilities to understand and compose formal and logical concepts, as claimed in Section 1.2.
    The paper states that PFMs can 'understand and compose complex semantics and concepts defined by nested or recursively built syntax structures' and can compose PAC hypotheses. This is a central premise for the claimed optimizations, but it is not proven or critically examined.
  • domain assumption The four optimization challenge areas (reasoning, accessibility, data quality, automation) are the key axes for improving data analysis with PFMs.
    Section 1.1 identifies these four challenges and organizes all later discussion around them. This is a modeling choice, not derived from evidence.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Revisiting Data Analysis with Pre-trained Foundation Models." pith.science (2026). https://pith.science/paper/2VUQAIRS

@misc{pith2026250101631,
  author       = {Pith},
  title        = {Pith review of: Revisiting Data Analysis with Pre-trained Foundation Models},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/2VUQAIRS}},
  note         = {Machine review of arXiv:2501.01631}
}
read the original abstract

Data analysis focuses on harnessing advanced statistics, programming, and machine learning techniques to extract valuable insights from vast datasets. An increasing volume and variety of research emerged, addressing datasets of diverse modalities, formats, scales, and resolutions across various industries. However, experienced data analysts often find themselves overwhelmed by intricate details in ad-hoc solutions or attempts to extract the semantics of grounded data properly. This makes it difficult to maintain and scale to more complex systems. Pre-trained foundation models (PFMs), grounded with a large amount of grounded data that previous data analysis methods can not fully understand, leverage complete statistics that combine reasoning of an admissible subset of results and statistical approximations by surprising engineering effects, to automate and enhance the analysis process. It pushes us to revisit data analysis to make better sense of data with PFMs. This paper provides a comprehensive review of systematic approaches to optimizing data analysis through the power of PFMs, while critically identifying the limitations of PFMs, to establish a roadmap for their future application in data analysis.

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

188 extracted references · 38 canonical work pages

  1. [7]

    Transformers as statisticians: Prov- able in-context learning with in-context algorithm selection

    Yu Bai et al. “Transformers as statisticians: Prov- able in-context learning with in-context algorithm selection”. In: Advances in neural information processing systems 36 (2024)

    2024

  2. [3]

    Physics of Language Models: Part 1, Context-Free Gram- mar

    Zeyuan Allen-Zhu and Yuanzhi Li. “Physics of Language Models: Part 1, Context-Free Gram- mar”. In: CoRR abs/2305.13673 (2023). doi: 10. 48550/ARXIV.2305.13673. arXiv: 2305.13673

    arXiv doi:10.48550/arxiv.2305.13673 2023

  3. [16]

    Language models are few- shot learners

    Tom Brown et al. “Language models are few- shot learners”. In: Advances in neural informa- tion processing systems 33 (2020), pp. 1877–1901

    2020

  4. [1]

    Generalization on the unseen, logic reasoning and degree curriculum

    Emmanuel Abbe et al. “Generalization on the unseen, logic reasoning and degree curriculum”. In: International Conference on Machine Learn- ing. PMLR. 2023, pp. 31–60

    2023

  5. [2]

    Interventional causal repre- sentation learning

    Kartik Ahuja et al. “Interventional causal repre- sentation learning”. In: International conference on machine learning. PMLR. 2023, pp. 372–407

    2023

  6. [4]

    Automated unit test improvement using large language models at meta

    Nadia Alshahwan et al. “Automated unit test improvement using large language models at meta”. In: Companion Proceedings of the 32nd ACM International Conference on the Foundations of Software Engineering. 2024, pp. 185–196

    2024

  7. [5]

    A survey of cross-validation procedures for model selection

    Sylvain Arlot and Alain Celisse. “A survey of cross-validation procedures for model selection”. In: (2010)

    2010

  8. [6]

    Infusing Lattice Symmetry Priors in Attention Mechanisms for Sample-Efficient Ab- stract Geometric Reasoning

    Mattia Atzeni, Mrinmaya Sachan, and Andreas Loukas. “Infusing Lattice Symmetry Priors in Attention Mechanisms for Sample-Efficient Ab- stract Geometric Reasoning”. In: International Conference on Machine Learning . 2023

    2023

Show all 188 references
  1. [8]

    Grounded Copilot: How Program- mers Interact with Code-Generating Models

    Shraddha Barke, Michael B. James, and Nadia Polikarpova. “Grounded Copilot: How Program- mers Interact with Code-Generating Models”. In: Proceedings of the ACM on Programming Lan- guages 7 (2022), pp. 85 –111

    2022

  2. [9]

    Methodologies for data qual- ity assessment and improvement

    Carlo Batini et al. “Methodologies for data qual- ity assessment and improvement”. In:ACM com- puting surveys (CSUR) 41.3 (2009), pp. 1–52

    2009

  3. [10]

    Neurosymbolic repair for low-code formula languages

    Rohan Bavishi et al. “Neurosymbolic repair for low-code formula languages”. In: Proceedings of the ACM on Programming Languages 6 (2022), pp. 1093 –1122

    2022

  4. [11]

    On the dangers of stochas- tic parrots: Can language models be too big?

    Emily M Bender et al. “On the dangers of stochas- tic parrots: Can language models be too big?” In: Proceedings of the 2021 ACM conference on fairness, accountability, and transparency. 2021, pp. 610–623

    2021

  5. [12]

    Graph2Tac: Online Representation Learning of Formal Math Con- cepts

    Lasse Blaauwbroek et al. “Graph2Tac: Online Representation Learning of Formal Math Con- cepts”. In: Forty-first International Conference on Machine Learning

  6. [13]

    A parameterized theory of PAC learn- Revisiting Data Analysis with Pre-trained Foundation Models 23 ing

    Cornelius Brand, Robert Ganian, and Kirill Si- monov. “A parameterized theory of PAC learn- Revisiting Data Analysis with Pre-trained Foundation Models 23 ing”. In: Proceedings of the AAAI Conference on Artificial Intelligence. Vol. 37. 6. 2023, pp. 6834– 6841

    2023

  7. [14]

    Statistical and computa- tional methods in data analysis

    Siegmund Brandt et al. Statistical and computa- tional methods in data analysis. 04. North-Holland Publishing Company Amsterdam, The Nether- lands: 1976

    1976

  8. [15]

    Practical Reliability Data Analy- sis for Non-Reliability Engineers

    Darcy Brooker. Practical Reliability Data Analy- sis for Non-Reliability Engineers. Artech House, 2020

    2020

  9. [17]

    Practical statistics for data scientists: 50+ es- sential concepts using R and Python

    Peter Bruce, Andrew Bruce, and Peter Gedeck. Practical statistics for data scientists: 50+ es- sential concepts using R and Python . O’Reilly Media, 2020

    2020

  10. [18]

    Ro- bustness of Nonlinear Representation Learning

    Simon Buchholz and Bernhard Sch¨ olkopf. “Ro- bustness of Nonlinear Representation Learning”. In: Forty-first International Conference on Ma- chine Learning

  11. [19]

    CoqPyt: Proof Navigation in Python in the Era of LLMs

    Pedro Carrott et al. “CoqPyt: Proof Navigation in Python in the Era of LLMs”. In: SIGSOFT FSE Companion. 2024

    2024

  12. [20]

    Why data sci- entists prefer glassbox machine learning: Algo- rithms, differential privacy, editing and bias mit- igation

    Rich Caruana and Harsha Nori. “Why data sci- entists prefer glassbox machine learning: Algo- rithms, differential privacy, editing and bias mit- igation”. In: Proceedings of the 28th ACM SIGKDD conference on knowledge discovery and data min- ing. 2022, pp. 4776–4777

    2022

  13. [21]

    SPIREX: Improving LLM-based relation extraction from RNA-focused scientific literature using graph machine learn- ing

    Emanuele Cavalleri et al. “SPIREX: Improving LLM-based relation extraction from RNA-focused scientific literature using graph machine learn- ing”. In: Proceedings of Workshops at the 50th International Conference on Very Large Data Bases, VLDB 2024, Guangzhou, China, August 26...

    2024

  14. [22]

    Binding Language Mod- els in Symbolic Languages

    Zhoujun Cheng et al. “Binding Language Mod- els in Symbolic Languages”. In: The Eleventh International Conference on Learning Represen- tations, ICLR 2023, Kigali, Rwanda, May 1-5,

    2023

  15. [23]

    Performance Optimiza- tion in the LLM World 2024

    Kingsum Chow et al. “Performance Optimiza- tion in the LLM World 2024”. In: Companion of the 15th ACM/SPEC International Conference on Performance Engineering. 2024, pp. 156–157

    2024

  16. [24]

    Relational database: A practi- cal foundation for productivity

    Edgar F Codd. “Relational database: A practi- cal foundation for productivity”. In: ACM Tur- ing award lectures. 2007, p. 1981

    2007

  17. [25]

    Large language models for compiler optimization

    Chris Cummins et al. “Large language models for compiler optimization”. In: arXiv preprint arXiv:2309.07062 (2023)

    2023 arXiv

  18. [26]

    Meta Large Language Model Compiler: Foundation Models of Com- piler Optimization

    Chris Cummins et al. “Meta Large Language Model Compiler: Foundation Models of Com- piler Optimization”. In: arXiv preprint arXiv:2407.02524 (2024)

    2024 arXiv

  19. [27]

    Good semi-supervised learn- ing that requires a bad gan

    Zihang Dai et al. “Good semi-supervised learn- ing that requires a bad gan”. In: Advances in neural information processing systems 30 (2017)

    2017

  20. [28]

    Compet- ing on analytics: Updated, with a new introduc- tion: The new science of winning

    Thomas Davenport and Jeanne Harris. Compet- ing on analytics: Updated, with a new introduc- tion: The new science of winning . Harvard Busi- ness Press, 2017

    2017

  21. [29]

    Language Modeling Is Compression

    Gr´ egoire Del´ etang et al. “Language Modeling Is Compression”. In: The Twelfth International Conference on Learning Representations, ICLR 2024, Vienna, Austria, May 7-11, 2024 . Open- Review.net, 2024

    2024

  22. [30]

    TURL: Table Understand- ing through Representation Learning

    Xiang Deng et al. “TURL: Table Understand- ing through Representation Learning”. In: Proc. VLDB Endow. 14.3 (2020), pp. 307–319. doi: 10.5555/3430915.3442430

    2020

  23. [31]

    LIDA: A Tool for Automatic Gen- eration of Grammar-Agnostic Visualizations and Infographics using Large Language Models

    Victor Dibia. “LIDA: A Tool for Automatic Gen- eration of Grammar-Agnostic Visualizations and Infographics using Large Language Models”. In: Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics: Sys- tem Demonstrations, ACL 2023, Toronto, Can...

    2023

  24. [32]

    Overfitting and undercomput- ing in machine learning

    Tom Dietterich. “Overfitting and undercomput- ing in machine learning”. In: ACM computing surveys (CSUR) 27.3 (1995), pp. 326–327

    1995

  25. [33]

    Large language models of code fail at completing code with potential bugs

    Tuan Dinh et al. “Large language models of code fail at completing code with potential bugs”. In: Advances in Neural Information Processing Sys- tems 36 (2024)

    2024

  26. [34]

    DeepJoin: Joinable Table Discovery with Pre-trained Language Models

    Yuyang Dong et al. “DeepJoin: Joinable Table Discovery with Pre-trained Language Models”. In: Proc. VLDB Endow. 16.10 (2023), pp. 2458–

    2023

  27. [35]

    Position: Compositional Generative Modeling: A Single Model is Not All You Need

    Yilun Du and Leslie Pack Kaelbling. “Position: Compositional Generative Modeling: A Single Model is Not All You Need”. In: Forty-first In- ternational Conference on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024 . Open- Review.net, 2024

    2024

  28. [36]

    Enhancing job recommen- dation through llm-based generative adversarial networks

    Yingpeng Du et al. “Enhancing job recommen- dation through llm-based generative adversarial networks”. In: Proceedings of the AAAI Confer- ence on Artificial Intelligence . Vol. 38. 8. 2024, pp. 8363–8371. 24 Chen Liang et al

    2024

  29. [37]

    On-device query intent prediction with lightweight LLMs to support ubiq- uitous conversations

    Mateusz Dubiel et al. “On-device query intent prediction with lightweight LLMs to support ubiq- uitous conversations”. In: Scientific Reports 14.1 (2024), p. 12731

    2024

  30. [38]

    DreamCoder: growing gener- alizable, interpretable knowledge with wake–sleep Bayesian program learning

    Kevin Ellis et al. “DreamCoder: growing gener- alizable, interpretable knowledge with wake–sleep Bayesian program learning”. In: Philosophical Transactions of the Royal Society A 381 (2020)

    2020

  31. [39]

    Semantics-aware Dataset Dis- covery from Data Lakes with Contextualized Column- based Representation Learning

    Grace Fan et al. “Semantics-aware Dataset Dis- covery from Data Lakes with Contextualized Column- based Representation Learning”. In:Proc. VLDB Endow. 16.7 (2023), pp. 1726–1739. doi: 10 . 14778/3587136.3587146

    2023

  32. [40]

    Combining Small Language Mod- els and Large Language Models for Zero-Shot NL2SQL

    Ju Fan et al. “Combining Small Language Mod- els and Large Language Models for Zero-Shot NL2SQL”. In: Proc. VLDB Endow. 17.11 (2024), pp. 2750–2763. doi: 10.14778/3681954.3681960

    2024

  33. [41]

    How large lan- guage models will disrupt data management

    Raul Castro Fernandez et al. “How large lan- guage models will disrupt data management”. In: Proceedings of the VLDB Endowment 16.11 (2023), pp. 3302–3309

    2023

  34. [42]

    Position: Relational Deep Learning-Graph Representation Learning on Re- lational Databases

    Matthias Fey et al. “Position: Relational Deep Learning-Graph Representation Learning on Re- lational Databases”. In: Forty-first International Conference on Machine Learning

  35. [43]

    CatSQL: Towards Real World Natural Language to SQL Applications

    Han Fu et al. “CatSQL: Towards Real World Natural Language to SQL Applications”. In:Proc. VLDB Endow. 16.6 (2023), pp. 1534–1547. doi: 10.14778/3583140.3583165

    2023

  36. [44]

    Text-to-SQL Empowered by Large Language Models: A Benchmark Evalua- tion

    Dawei Gao et al. “Text-to-SQL Empowered by Large Language Models: A Benchmark Evalua- tion”. In: Proc. VLDB Endow. 17.5 (2024), pp. 1132–

    2024

  37. [45]

    Retrieval-augmented gener- ation for large language models: A survey

    Yunfan Gao et al. “Retrieval-augmented gener- ation for large language models: A survey”. In: arXiv preprint arXiv:2312.10997 (2023)

    2023 arXiv

  38. [46]

    Using experts to develop a supply chain matu- rity model in Mexico

    Heriberto Garcia Reyes and Ronald Giachetti. “Using experts to develop a supply chain matu- rity model in Mexico”. In: Supply Chain Man- agement: An International Journal 15.6 (2010), pp. 415–424

    2010

  39. [47]

    LLM-PBC: Logic Learn- ing Machine-based explainable rules accurately stratify the genetic risk of Primary Biliary Cholan- gitis

    Alessio Gerussi et al. “LLM-PBC: Logic Learn- ing Machine-based explainable rules accurately stratify the genetic risk of Primary Biliary Cholan- gitis”. In: Journal of Personalized Medicine 12.10 (2022), p. 1587

    2022

  40. [48]

    LILO: Learning Interpretable Libraries by Compressing and Documenting Code

    Gabriel Grand et al. “LILO: Learning Interpretable Libraries by Compressing and Documenting Code”. In: The Twelfth International Conference on Learn- ing Representations, ICLR 2024, Vienna, Aus- tria, May 7-11, 2024 . OpenReview.net, 2024

    2024

  41. [49]

    Why do tree-based models still outper- form deep learning on typical tabular data?

    L´ eo Grinsztajn, Edouard Oyallon, and Ga¨ el Varo- quaux. “Why do tree-based models still outper- form deep learning on typical tabular data?” In: Advances in neural information processing sys- tems 35 (2022), pp. 507–520

    2022

  42. [50]

    Few-shot text-to-sql translation using structure and content prompt learning

    Zihui Gu et al. “Few-shot text-to-sql translation using structure and content prompt learning”. In: Proceedings of the ACM on Management of Data 1.2 (2023), pp. 1–28

    2023

  43. [51]

    An intro- duction to variable and feature selection

    Isabelle Guyon and Andr´ e Elisseeff. “An intro- duction to variable and feature selection”. In: Journal of machine learning research 3.Mar (2003), pp. 1157–1182

    2003

  44. [52]

    Interpreting Equivariant Repre- sentations

    Andreas Abildtrup Hansen, Anna Calissano, and Aasa Feragen. “Interpreting Equivariant Repre- sentations”. In: Forty-first International Confer- ence on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024. OpenReview.net, 2024

    2024

  45. [53]

    The elements of statistical learning: data mining, inference, and prediction

    Trevor Hastie et al. The elements of statistical learning: data mining, inference, and prediction. Vol. 2. Springer, 2009

    2009

  46. [54]

    Optimizing Video Selection LIMIT Queries With Commonsense Knowledge

    Wenjia He et al. “Optimizing Video Selection LIMIT Queries With Commonsense Knowledge”. In: Proc. VLDB Endow. 17.7 (2024), pp. 1751–

    2024

  47. [55]

    Declarative language design for interactive visualization

    Jeffrey Heer and Michael Bostock. “Declarative language design for interactive visualization”. In: IEEE transactions on visualization and computer graphics 16.6 (2010), pp. 1149–1156

    2010

  48. [56]

    Large Language Models for Automated Data Science: Introducing CAAFE for Context- Aware Automated Feature Engineering

    Noah Hollmann, Samuel G. M¨ uller, and Frank Hutter. “Large Language Models for Automated Data Science: Introducing CAAFE for Context- Aware Automated Feature Engineering”. In:Neu- ral Information Processing Systems . 2023

    2023

  49. [57]

    What’s Left? Concept Ground- ing with Logic-Enhanced Foundation Models

    Joy Hsu et al. “What’s Left? Concept Ground- ing with Logic-Enhanced Foundation Models”. In: Advances in Neural Information Processing Systems 36: Annual Conference on Neural Infor- mation Processing Systems 2023, NeurIPS 2023, New Orleans, LA, USA, December 10 - 16, 2023. Ed....

    2023

  50. [58]

    Amortizing intractable in- ference in large language models

    Edward J Hu et al. “Amortizing intractable in- ference in large language models”. In:The Twelfth International Conference on Learning Represen- tations

  51. [59]

    A survey of knowledge en- hanced pre-trained language models

    Linmei Hu et al. “A survey of knowledge en- hanced pre-trained language models”. In: IEEE Transactions on Knowledge and Data Engineer- ing (2023)

    2023

  52. [60]

    InfiAgent-DABench: Evaluat- ing Agents on Data Analysis Tasks

    Xueyu Hu et al. “InfiAgent-DABench: Evaluat- ing Agents on Data Analysis Tasks”. In: Forty- first International Conference on Machine Learn- ing, ICML 2024, Vienna, Austria, July 21-27,

    2024

  53. [61]

    KOSA: KO enhanced salary analytics based on knowledge graph and LLM capabilities

    Fei Huang et al. “KOSA: KO enhanced salary analytics based on knowledge graph and LLM capabilities”. In: 2023 IEEE International Con- ference on Data Mining Workshops (ICDMW) . IEEE. 2023, pp. 499–505

    2023

  54. [62]

    DAMA-DMBOK: Data man- agement body of knowledge

    Dama International. DAMA-DMBOK: Data man- agement body of knowledge . Technics Publica- tions, LLC, 2017

    2017

  55. [63]

    Deep Indexed Active Learning for Matching Het- erogeneous Entity Representations

    Arjit Jain, Sunita Sarawagi, and Prithviraj Sen. “Deep Indexed Active Learning for Matching Het- erogeneous Entity Representations”. In: Proc. VLDB Endow. 15.1 (2021), pp. 31–45. doi: 10. 14778/3485450.3485455

    2021

  56. [64]

    Jigsaw: Large language mod- els meet program synthesis

    Naman Jain et al. “Jigsaw: Large language mod- els meet program synthesis”. In: Proceedings of the 44th International Conference on Software Engineering. 2022, pp. 1219–1231

    2022

  57. [65]

    R2E: Turning any Github Repository into a Programming Agent Environ- ment

    Naman Jain et al. “R2E: Turning any Github Repository into a Programming Agent Environ- ment”. In: Forty-first International Conference on Machine Learning . 2024

    2024

  58. [66]

    Counterexample guided inductive synthesis using large language models and satisfiability solving

    Sumit Kumar Jha et al. “Counterexample guided inductive synthesis using large language models and satisfiability solving”. In: MILCOM 2023- 2023 IEEE Military Communications Conference (MILCOM). IEEE. 2023, pp. 944–949

    2023

  59. [67]

    Survey of Hallucination in Natu- ral Language Generation

    Ziwei Ji et al. “Survey of Hallucination in Natu- ral Language Generation”. In: ACM Computing Surveys 55 (2022), pp. 1 –38

    2022

  60. [68]

    The 3W Model and Alge- bra for Unified Data Mining

    Theodore Johnson, Laks V. S. Lakshmanan, and Raymond T. Ng. “The 3W Model and Alge- bra for Unified Data Mining”. In: VLDB 2000, Proceedings of 26th International Conference on Very Large Data Bases, September 10-14, 2000, Cairo, Egypt. Ed. by Amr El Abbadi et al. Mor- gan Kau...

    2000

  61. [69]

    Reinforcement learning: A survey

    Leslie Pack Kaelbling, Michael L Littman, and Andrew W Moore. “Reinforcement learning: A survey”. In: Journal of artificial intelligence re- search 4 (1996), pp. 237–285

    1996

  62. [70]

    A Survey of Reinforce- ment Learning from Human Feedback

    Timo Kaufmann et al. “A Survey of Reinforce- ment Learning from Human Feedback”. In:ArXiv abs/2312.14925 (2023)

    2023

  63. [71]

    CHORUS: Foundation Mod- els for Unified Data Discovery and Exploration

    Moe Kayali et al. “CHORUS: Foundation Mod- els for Unified Data Discovery and Exploration”. In: Proc. VLDB Endow. 17.8 (2024), pp. 2104–

    2024

  64. [72]

    Information quality: The potential of data and analytics to generate knowledge

    Ron S Kenett and Galit Shmueli. Information quality: The potential of data and analytics to generate knowledge. John Wiley & Sons, 2016

    2016

  65. [73]

    PAC Prediction Sets for Large Language Models of Code

    Adam Khakhar, Stephen Mell, and Osbert Bas- tani. “PAC Prediction Sets for Large Language Models of Code”. In: International Conference on Machine Learning . 2023

    2023

  66. [74]

    PAC prediction sets for large language models of code

    Adam Khakhar, Stephen Mell, and Osbert Bas- tani. “PAC prediction sets for large language models of code”. In: International Conference on Machine Learning . PMLR. 2023, pp. 16237– 16249

    2023

  67. [75]

    Object identity

    Setrag N Khoshafian and George P Copeland. “Object identity”. In: ACM SIGPLAN Notices 21.11 (1986), pp. 406–416

    1986

  68. [76]

    Cicero: A declarative gram- mar for responsive visualization

    Hyeok Kim et al. “Cicero: A declarative gram- mar for responsive visualization”. In: Proceed- ings of the 2022 CHI Conference on Human Fac- tors in Computing Systems . 2022, pp. 1–15

    2022

  69. [77]

    Filling in the Gaps: LLM- Based Structured Data Generation from Semi- Structured Scientific Data

    Hanbum Ko et al. “Filling in the Gaps: LLM- Based Structured Data Generation from Semi- Structured Scientific Data”. In: ICML 2024 AI for Science Workshop . 2024

    2024

  70. [78]

    LLM-based and Retrieval- Augmented Control Code Generation

    Heiko Koziolek et al. “LLM-based and Retrieval- Augmented Control Code Generation”. In:LLM4CODE@ICSE. 2024, pp. 22–29. doi: 10.1145/3643795.3648384

    2024

  71. [79]

    DS-1000: A natural and re- liable benchmark for data science code genera- tion

    Yuhang Lai et al. “DS-1000: A natural and re- liable benchmark for data science code genera- tion”. In: International Conference on Machine Learning. PMLR. 2023, pp. 18319–18345

    2023

  72. [81]

    Modern data analysis

    Robert L Launer and Andrew F Siegel. Modern data analysis . Academic Press, 2014

    2014

  73. [82]

    Language Models as Controlled Natural Language Semantic Parsers for Knowledge Graph Question Answering

    Jens Lehmann, S´ ebastien Ferr´ e, and Sahar Vah- dati. “Language Models as Controlled Natural Language Semantic Parsers for Knowledge Graph Question Answering”. In: European Conference on Artificial Intelligence . 2023

    2023

  74. [83]

    The Dawn of Natural Language to SQL: Are We Fully Ready? [Experiment, Anal- ysis & Benchmark ]

    Boyan Li et al. “The Dawn of Natural Language to SQL: Are We Fully Ready? [Experiment, Anal- ysis & Benchmark ]”. In: Proc. VLDB Endow. 17.11 (2024), pp. 3318–3331. doi: 10 . 14778 / 3681954.3682003

    2024

  75. [84]

    Resdsql: Decoupling schema linking and skeleton parsing for text-to-sql

    Haoyang Li et al. “Resdsql: Decoupling schema linking and skeleton parsing for text-to-sql”. In: Proceedings of the AAAI Conference on Artifi- cial Intelligence . Vol. 37. 11. 2023, pp. 13067– 13075

    2023

  76. [85]

    Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls

    Jinyang Li et al. “Can llm already serve as a database interface? a big bench for large-scale database grounded text-to-sqls”. In: Advances in Neural Information Processing Systems 36 (2024)

    2024

  77. [86]

    Is Programming by Example solved by LLMs?

    Wen-Ding Li and Kevin Ellis. “Is Programming by Example solved by LLMs?” In:ArXiv abs/2406.08316 (2024). 26 Chen Liang et al

    2024 arXiv

  78. [87]

    Towards Efficient Data Wrangling with LLMs using Code Generation

    Xue Li and Till D¨ ohmen. “Towards Efficient Data Wrangling with LLMs using Code Generation”. In: Proceedings of the Eighth Workshop on Data Management for End-to-End Machine Learning, DEEM 2024, Santiago, AA, Chile, 9 June 2024 . ACM, 2024, pp. 62–66. doi: 10.1145/3650203. 3663334

    2024 doi

  79. [88]

    Towards Efficient Data Wrangling with LLMs using Code Generation

    Xue Li and Till D¨ ohmen. “Towards Efficient Data Wrangling with LLMs using Code Generation”. In: Proceedings of the Eighth Workshop on Data Management for End-to-End Machine Learning. 2024, pp. 62–66

    2024

  80. [89]

    Deep Entity Matching with Pre-Trained Language Models

    Yuliang Li et al. “Deep Entity Matching with Pre-Trained Language Models”. In: Proc. VLDB Endow. 14.1 (2020), pp. 50–60. doi: 10.14778/ 3421424.3421431

    2020

  81. [90]

    Flexkbqa: A flexible llm-powered framework for few-shot knowledge base question answering

    Zhenyu Li et al. “Flexkbqa: A flexible llm-powered framework for few-shot knowledge base question answering”. In: Proceedings of the AAAI Confer- ence on Artificial Intelligence . Vol. 38. 17. 2024, pp. 18608–18616

    2024

  82. [91]

    Foundation Models for Time Series Analysis: A Tutorial and Survey

    Yuxuan Liang et al. “Foundation Models for Time Series Analysis: A Tutorial and Survey”. In:Pro- ceedings of the 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, KDD 2024, Barcelona, Spain, August 25-29, 2024. Ed. by Ricardo Baeza-Yates and Francesco Bonchi. ...

    2024 doi

  83. [92]

    Statis- tical analysis with missing data

    Roderick JA Little and Donald B Rubin. Statis- tical analysis with missing data . Vol. 793. John Wiley & Sons, 2019

    2019

  84. [93]

    JarviX: A LLM no code platform for tabular data analysis and optimiza- tion

    Shang-Ching Liu et al. “JarviX: A LLM no code platform for tabular data analysis and optimiza- tion”. In: Proceedings of the 2023 Conference on Empirical Methods in Natural Language Process- ing: Industry Track. 2023, pp. 622–630

    2023

  85. [94]

    Enhancing Large Language Mod- els with Multimodality and Knowledge Graphs for Hallucination-free Open-set Object Recogni- tion

    Xinfu Liu et al. “Enhancing Large Language Mod- els with Multimodality and Knowledge Graphs for Hallucination-free Open-set Object Recogni- tion”. In: Proceedings of Workshops at the 50th International Conference on Very Large Data Bases, VLDB 2024, Guangzhou, China, August 26...

    2024

  86. [95]

    SAIE Framework: Support Alone Isn’t Enough - Advancing LLM Training with Ad- versarial Remarks

    Mengsay Loem, Masahiro Kaneko, and Naoaki Okazaki. “SAIE Framework: Support Alone Isn’t Enough - Advancing LLM Training with Ad- versarial Remarks”. In: European Conference on Artificial Intelligence. 2023

    2023

  87. [96]

    Proof Automation with Large Language Mod- els

    Minghai Lu, Benjamin Delaware, and Tianyi Zhang. “Proof Automation with Large Language Mod- els”. In: ArXiv abs/2409.14274 (2024)

    2024 arXiv

  88. [97]

    Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity

    Yao Lu et al. “Fantastically Ordered Prompts and Where to Find Them: Overcoming Few-Shot Prompt Order Sensitivity”. In:ArXiv abs/2104.08786 (2021)

    2021 arXiv

  89. [99]

    Automatic programming: Large language models and beyond

    Michael R Lyu et al. “Automatic programming: Large language models and beyond”. In: ACM Transactions on Software Engineering and Method- ology (2024)

    2024

  90. [100]

    InsightPilot: An LLM- empowered automated data exploration system

    Pingchuan Ma et al. “InsightPilot: An LLM- empowered automated data exploration system”. In: Proceedings of the 2023 Conference on Em- pirical Methods in Natural Language Processing: System Demonstrations. 2023, pp. 346–352

    2023

  91. [101]

    Foundation Models for Video Understanding: A Survey

    Neelu Madan et al. “Foundation Models for Video Understanding: A Survey”. In:CoRR abs/2405.03770 (2024). doi: 10 . 48550 / ARXIV . 2405 . 03770. arXiv: 2405.03770

    2024 doi arXiv

  92. [102]

    Learning Instance-Specific Aug- mentations by Capturing Local Invariances

    Ning Miao et al. “Learning Instance-Specific Aug- mentations by Capturing Local Invariances”. In: International Conference on Machine Learning . 2022

    2022

  93. [103]

    Learning deductive reasoning from synthetic corpus based on formal logic

    Terufumi Morishita et al. “Learning deductive reasoning from synthetic corpus based on formal logic”. In: International Conference on Machine Learning. PMLR. 2023, pp. 25254–25274

    2023

  94. [104]

    LLM Guided Evolution - The Automation of Models Advancing Models

    Clint Morris, Michael Jurado, and Jason Zutty. “LLM Guided Evolution - The Automation of Models Advancing Models”. In: Proceedings of the Genetic and Evolutionary Computation Con- ference, GECCO 2024, Melbourne, VIC, Aus- tralia, July 14-18, 2024 . Ed. by Xiaodong Li and Julia...

    2024

  95. [105]

    Making sense of data: a practi- cal guide to exploratory data analysis and data mining

    Glenn J Myatt. Making sense of data: a practi- cal guide to exploratory data analysis and data mining. John Wiley & Sons, 2007

    2007

  96. [106]

    Using an llm to help with code understanding

    Daye Nam et al. “Using an llm to help with code understanding”. In: Proceedings of the IEEE/ACM 46th International Conference on Software En- gineering. 2024, pp. 1–13

    2024

  97. [107]

    Optimized Feature Gener- ation for Tabular Data via LLMs with Decision Tree Reasoning

    Jaehyun Nam et al. “Optimized Feature Gener- ation for Tabular Data via LLMs with Decision Tree Reasoning”. In:arXiv preprint arXiv:2406.08527 (2024)

    2024 arXiv

  98. [108]

    Can Foundation Mod- els Wrangle Your Data?

    Avanika Narayan et al. “Can Foundation Mod- els Wrangle Your Data?” In: Proc. VLDB En- dow. 16.4 (2022), pp. 738–746. doi: 10.14778/ 3574245.3574258. Revisiting Data Analysis with Pre-trained Foundation Models 27

    2022

  99. [109]

    Generalization Bounds via Convex Analysis

    Gergely Neu and G´ abor Lugosi. “Generalization Bounds via Convex Analysis”. In: Annual Con- ference Computational Learning Theory. 2022

    2022

  100. [110]

    Mallet: SQL Dialect Translation with LLM Rule Gener- ation

    Amadou Latyr Ngom and Tim Kraska. “Mallet: SQL Dialect Translation with LLM Rule Gener- ation”. In: Proceedings of the Seventh Interna- tional Workshop on Exploiting Artificial Intelli- gence Techniques for Data Management . 2024, pp. 1–5

    2024

  101. [111]

    IterClean: An Iterative Data Clean- ing Framework with Large Language Models

    Wei Ni et al. “IterClean: An Iterative Data Clean- ing Framework with Large Language Models”. In: Proceedings of the ACM Turing Award Cel- ebration Conference-China 2024. 2024, pp. 100– 105

    2024

  102. [112]

    Knowledge Graph Efficient Construction: Embedding Chain-of-Thought into LLMs

    Jixuan Nie et al. “Knowledge Graph Efficient Construction: Embedding Chain-of-Thought into LLMs”. In: Proceedings of Workshops at the 50th International Conference on Very Large Data Bases, VLDB 2024, Guangzhou, China, August 26-30, 2024 . VLDB.org, 2024

    2024

  103. [113]

    Handbook of statistical analysis and data mining applications

    Robert Nisbet, John Elder, and Gary D Miner. Handbook of statistical analysis and data mining applications. Academic press, 2009

    2009

  104. [114]

    DTT: An Example-Driven Tabular Transformer for Join- ability by Leveraging Large Language Models

    Arash Dargahi Nobari and Davood Rafiei. “DTT: An Example-Driven Tabular Transformer for Join- ability by Leveraging Large Language Models”. In: Proceedings of the ACM on Management of Data 2 (2023), pp. 1 –24

    2023

  105. [115]

    Data quality: the accuracy dimen- sion

    Jack E Olson. Data quality: the accuracy dimen- sion. Elsevier, 2003

    2003

  106. [116]

    ALT-GEN: Benchmarking Table Union Search using Large Language Mod- els

    Koyena Pal et al. “ALT-GEN: Benchmarking Table Union Search using Large Language Mod- els”. In: Proceedings of Workshops at the 50th International Conference on Very Large Data Bases, VLDB 2024, Guangzhou, China, August 26-30, 2024 . VLDB.org, 2024

    2024

  107. [117]

    Schema Matching with Large Language Models: an Experimental Study

    Marcel Parciak et al. “Schema Matching with Large Language Models: an Experimental Study”. In: Proceedings of Workshops at the 50th Inter- national Conference on Very Large Data Bases, VLDB 2024, Guangzhou, China, August 26-30,

    2024

  108. [118]

    The book of why: the new science of cause and effect

    Judea Pearl and Dana Mackenzie. The book of why: the new science of cause and effect . Basic books, 2018

    2018

  109. [119]

    Can Large Language Models Reason about Program Invariants?

    Kexin Pei et al. “Can Large Language Models Reason about Program Invariants?” In: Interna- tional Conference on Machine Learning . 2023

    2023

  110. [120]

    Can large language models reason about program invariants?

    Kexin Pei et al. “Can large language models reason about program invariants?” In: Interna- tional Conference on Machine Learning. PMLR. 2023, pp. 27496–27520

    2023

  111. [121]

    Self-supervised and inter- pretable data cleaning with sequence generative adversarial networks

    Jinfeng Peng et al. “Self-supervised and inter- pretable data cleaning with sequence generative adversarial networks”. In:Proceedings of the VLDB Endowment 16.3 (2022), pp. 433–446

    2022

  112. [122]

    Neural oblivious decision ensembles for deep learning on tabular data

    Sergei Popov, Stanislav Morozov, and Artem Babenko. “Neural oblivious decision ensembles for deep learning on tabular data”. In:arXiv preprint arXiv:1909.06312 (2019)

    2019 arXiv

  113. [123]

    Data Science for Business: What you need to know about data mining and data-analytic thinking

    F Provost. Data Science for Business: What you need to know about data mining and data-analytic thinking. Vol. 355. O’Reilly Media, Inc, 2013

    2013

  114. [124]

    CleanAgent: Au- tomating Data Standardization with LLM-based Agents

    Danrui Qi and Jiannan Wang. “CleanAgent: Au- tomating Data Standardization with LLM-based Agents”. In: arXiv preprint arXiv:2403.08291 (2024)

    2024 arXiv

  115. [125]

    Position: Understanding LLMs Requires More Than Statistical General- ization

    Patrik Reizinger et al. “Position: Understanding LLMs Requires More Than Statistical General- ization”. In: Forty-first International Conference on Machine Learning

  116. [126]

    doi: 10.18653/V1/2023.ACL-DEMO.11

    2023 doi

  117. [127]

    AutoML: A systematic re- view on automated machine learning with neural architecture search

    Imrus Salehin et al. “AutoML: A systematic re- view on automated machine learning with neural architecture search”. In: Journal of Information and Intelligence 2.1 (2024), pp. 52–81

    2024

  118. [128]

    GizaML: A Collaborative Meta-learning Based Framework Using LLM For Automated Time-Series Forecasting

    Esraa Sayed et al. “GizaML: A Collaborative Meta-learning Based Framework Using LLM For Automated Time-Series Forecasting.” In:EDBT. 2024, pp. 830–833

    2024

  119. [129]

    An empirical evaluation of using large language models for automated unit test generation

    Max Sch¨ afer et al. “An empirical evaluation of using large language models for automated unit test generation”. In: IEEE Transactions on Soft- ware Engineering (2023)

    2023

  120. [130]

    Automated theorem prov- ing in software engineering

    Johann M Schumann. Automated theorem prov- ing in software engineering . Springer Science & Business Media, 2013

    2013

  121. [131]

    Synthetic prompting: Gen- erating chain-of-thought demonstrations for large language models

    Zhihong Shao et al. “Synthetic prompting: Gen- erating chain-of-thought demonstrations for large language models”. In: International Conference on Machine Learning . PMLR. 2023, pp. 30706– 30775

    2023

  122. [132]

    Large Language Models Can Be Easily Distracted by Irrelevant Context

    Freda Shi et al. “Large Language Models Can Be Easily Distracted by Irrelevant Context”. In: International Conference on Machine Learning . 2023

    2023

  123. [133]

    Position: Understanding LLMs Requires More Than Statistical Gener- alization

    Patrik Reizinger et al. “Position: Understanding LLMs Requires More Than Statistical Gener- alization”. In: Forty-first International Confer- ence on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024. OpenReview.net, 2024

    2024

  124. [134]

    Predictive analytics in information systems research

    Galit Shmueli and Otto R Koppius. “Predictive analytics in information systems research”. In: MIS quarterly (2011), pp. 553–572

    2011

  125. [135]

    Augmenting interpretable models with large language models during train- ing

    Chandan Singh et al. “Augmenting interpretable models with large language models during train- ing”. In: Nature Communications 14.1 (2023), p. 7913

    2023

  126. [136]

    Latent Logic Tree Extraction for Event Sequence Explanation from LLMs

    Zitao Song et al. “Latent Logic Tree Extraction for Event Sequence Explanation from LLMs”. In: Forty-first International Conference on Ma- chine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024 . OpenReview.net, 2024

    2024

  127. [137]

    Joint prompt opti- mization of stacked llms using variational infer- ence

    Alessandro Sordoni et al. “Joint prompt opti- mization of stacked llms using variational infer- ence”. In: Advances in Neural Information Pro- cessing Systems 36 (2024)

    2024

  128. [138]

    Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins

    Sahaana Suri et al. “Ember: No-Code Context Enrichment via Similarity-Based Keyless Joins”. In: Proc. VLDB Endow. 15.3 (2021), pp. 699–

    2021

  129. [139]

    Are Language Models Ac- tually Useful for Time Series Forecasting?

    Mingtian Tan et al. “Are Language Models Ac- tually Useful for Time Series Forecasting?” In: arXiv preprint arXiv:2406.16964 (2024)

    2024 arXiv

  130. [140]

    A declarative grammar of flexible volume visu- alization pipelines

    Min Shih, Charles Rozhon, and Kwan-Liu Ma. “A declarative grammar of flexible volume visu- alization pipelines”. In: IEEE Transactions on Visualization and Computer Graphics 25.1 (2018), pp. 1050–1059. 28 Chen Liang et al

    2018

  131. [141]

    From Natural Language Processing to Neural Databases

    James Thorne et al. “From Natural Language Processing to Neural Databases”. In:Proc. VLDB Endow. 14.6 (2021), pp. 1033–1039. doi: 10 . 14778/3447689.3447706

    2021

  132. [142]

    Large language models should be used as scientific reasoning engines, not knowl- edge databases

    Daniel Truhn, Jorge S Reis-Filho, and Jakob Nikolas Kather. “Large language models should be used as scientific reasoning engines, not knowl- edge databases”. In: Nature medicine 29.12 (2023), pp. 2983–2984

    2023

  133. [143]

    CodexDB: Synthesizing code for query processing from natural language in- structions using GPT-3 Codex

    Immanuel Trummer. “CodexDB: Synthesizing code for query processing from natural language in- structions using GPT-3 Codex”. In: Proceedings of the VLDB Endowment 15.11 (2022), pp. 2921– 2928

    2022

  134. [144]

    The Case for NLP-Enhanced Database Tuning: Towards Tuning Tools that

    Immanuel Trummer. “The Case for NLP-Enhanced Database Tuning: Towards Tuning Tools that ”Read the Manual””. In: Proc. VLDB Endow. 14.7 (2021), pp. 1159–1165. doi: 10 . 14778 / 3450980.3450984

    2021

  135. [145]

    On computable numbers, with an application to the Entscheidungs problem

    A Turing. “On computable numbers, with an application to the Entscheidungs problem”. In: Proceedings of the London Mathematical Society Series/2 (42) (1936), pp. 230–42

    1936

  136. [146]

    Data Management: a gentle in- troduction: Balancing theory and practice

    Bas Van Gils. Data Management: a gentle in- troduction: Balancing theory and practice . Van Haren, 2020

    2020

  137. [147]

    Rethink- ing statistical learning theory: learning using sta- tistical invariants

    Vladimir Vapnik and Rauf Izmailov. “Rethink- ing statistical learning theory: learning using sta- tistical invariants”. In: Machine Learning 108.3 (2019), pp. 381–423

    2019

  138. [148]

    World- Coder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment

    Hao Tang, Darren Key, and Kevin Ellis. “World- Coder, a Model-Based LLM Agent: Building World Models by Writing Code and Interacting with the Environment”. In:ArXiv abs/2402.12275 (2024)

    2024 arXiv

  139. [149]

    Editable ma- chine learning models? A rule-based framework for user studies of explainability

    Stanislav Voj ´ ıˇ r and Tom´ aˇ s Kliegr. “Editable ma- chine learning models? A rule-based framework for user studies of explainability”. In: Advances in Data Analysis and Classification 14.4 (2020), pp. 785–799

    2020

  140. [150]

    Towards parameter-efficient automation of data wrangling tasks with prefix-tuning

    David Vos, Till D¨ ohmen, and Sebastian Schel- ter. “Towards parameter-efficient automation of data wrangling tasks with prefix-tuning”. In:NeurIPS 2022 First Table Representation Workshop. 2022

    2022

  141. [151]

    Large Language Model for Ge- ometric Algebra: A Preliminary Attempt

    Jian Wang et al. “Large Language Model for Ge- ometric Algebra: A Preliminary Attempt”. In: Computer Graphics International Conference. Springer. 2023, pp. 237–249

    2023

  142. [152]

    Solo: Data Discovery Using Natural Language Ques- tions Via A Self-Supervised Approach

    Qiming Wang and Raul Castro Fernandez. “Solo: Data Discovery Using Natural Language Ques- tions Via A Self-Supervised Approach”. In: Pro- ceedings of the ACM on Management of Data 1 (2023), pp. 1 –27

    2023

  143. [153]

    Beyond accuracy: What data quality means to data con- sumers

    Richard Y Wang and Diane M Strong. “Beyond accuracy: What data quality means to data con- sumers”. In: Journal of management informa- tion systems 12.4 (1996), pp. 5–33

    1996

  144. [154]

    TheoremLlama: Transform- ing General-Purpose LLMs into Lean4 Experts

    Ruida Wang et al. “TheoremLlama: Transform- ing General-Purpose LLMs into Lean4 Experts”. In: ArXiv (2024)

    2024

  145. [155]

    Knowledge Editing for Large Language Models: A Survey

    Song Wang et al. “Knowledge Editing for Large Language Models: A Survey”. In:ArXiv abs/2310.16218 (2023)

    2023 arXiv

  146. [156]

    Hy- brid LLM/Rule-based Approaches to Business Insights Generation from Structured Data

    Aliaksei Vertsel and Mikhail Rumiantsau. “Hy- brid LLM/Rule-based Approaches to Business Insights Generation from Structured Data”. In: arXiv preprint arXiv:2404.15604 (2024)

    2024 arXiv

  147. [157]

    G-tuning: Improving generalization of pre-trained language models with generative adversarial net- work

    Rongxiang Weng, Wen Sen Cheng, and Min Zhang. “G-tuning: Improving generalization of pre-trained language models with generative adversarial net- work”. In: Findings of the Association for Com- putational Linguistics: ACL 2023. 2023, pp. 4747– 4755

    2023

  148. [158]

    Repoformer: Selective Retrieval for Repository-Level Code Completion

    Di Wu et al. “Repoformer: Selective Retrieval for Repository-Level Code Completion”. In: Forty- first International Conference on Machine Learn- ing, ICML 2024, Vienna, Austria, July 21-27,

    2024

  149. [159]

    Multimodal large language models: A survey

    Jiayang Wu et al. “Multimodal large language models: A survey”. In: 2023 IEEE International Conference on Big Data (BigData). IEEE. 2023, pp. 2247–2256

    2023

  150. [160]

    Continual learning for large language models: A survey

    Tongtong Wu et al. “Continual learning for large language models: A survey”. In: arXiv preprint arXiv:2402.01364 (2024)

    2024 arXiv

  151. [161]

    Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond

    Jingfeng Yang et al. “Harnessing the Power of LLMs in Practice: A Survey on ChatGPT and Beyond”. In: ACM Trans. Knowl. Discov. Data 18.6 (2024), 160:1–160:32. doi: 10.1145/3649506

    2024 doi

  152. [162]

    Leandojo: Theorem proving with retrieval-augmented language models

    Kaiyu Yang et al. “Leandojo: Theorem proving with retrieval-augmented language models”. In: Advances in Neural Information Processing Sys- tems 36 (2024)

    2024

  153. [163]

    Give us the facts: Enhancing large language models with knowledge graphs for fact-aware language modeling

    Linyao Yang et al. “Give us the facts: Enhancing large language models with knowledge graphs for fact-aware language modeling”. In:IEEE Trans- actions on Knowledge and Data Engineering (2024)

    2024

  154. [164]

    Executable Code Actions Elicit Better LLM Agents

    Xingyao Wang et al. “Executable Code Actions Elicit Better LLM Agents”. In: Forty-first Inter- national Conference on Machine Learning, ICML 2024, Vienna, Austria, July 21-27, 2024 . Open- Review.net, 2024

    2024

  155. [165]

    Tree of thoughts: Deliberate problem solving with large language models

    Shunyu Yao et al. “Tree of thoughts: Deliberate problem solving with large language models”. In: Advances in Neural Information Processing Sys- tems 36 (2024)

    2024

  156. [166]

    Satlm: Satisfiability-aided language models using declarative prompting

    Xi Ye et al. “Satlm: Satisfiability-aided language models using declarative prompting”. In: Ad- vances in Neural Information Processing Sys- tems 36 (2024)

    2024

  157. [167]

    Revisiting Data Analysis with Pre-trained Foundation Models 29

    OpenReview.net, 2024. Revisiting Data Analysis with Pre-trained Foundation Models 29

    2024

  158. [168]

    Semi-supervised learning with deep generative models for asset failure pre- diction

    Andre S Yoon et al. “Semi-supervised learning with deep generative models for asset failure pre- diction”. In: arXiv preprint arXiv:1709.00845 (2017)

    2017 arXiv

  159. [169]

    On the power of foundation mod- els

    Yang Yuan. “On the power of foundation mod- els”. In: International Conference on Machine Learning. PMLR. 2023, pp. 40519–40530

    2023

  160. [170]

    On the Power of Foundation Mod- els

    Yang Yuan. “On the Power of Foundation Mod- els”. In: International Conference on Machine Learning, ICML 2023, 23-29 July 2023, Hon- olulu, Hawaii, USA . Ed. by Andreas Krause et al. Vol. 202. Proceedings of Machine Learning Research. PMLR, 2023, pp. 40519–40530

    2023

  161. [171]

    Pre-trained Embed- dings for Entity Resolution: An Experimental Analysis

    Alexandros Zeakis et al. “Pre-trained Embed- dings for Entity Resolution: An Experimental Analysis”. In: Proc. VLDB Endow. 16.9 (2023), pp. 2225–2238. doi: 10.14778/3598581.3598594

    2023

  162. [172]

    Scaling vision transform- ers

    Xiaohua Zhai et al. “Scaling vision transform- ers”. In: Proceedings of the IEEE/CVF confer- ence on computer vision and pattern recognition. 2022, pp. 12104–12113

    2022

  163. [173]

    Deep learning for practical im- age recognition: Case study on kaggle competi- tions

    Xulei Yang et al. “Deep learning for practical im- age recognition: Case study on kaggle competi- tions”. In: Proceedings of the 24th ACM SIGKDD international conference on knowledge discovery & data mining . 2018, pp. 923–931

    2018

  164. [174]

    Experimental Analysis of Large-scale Learnable Vector Storage Compres- sion

    Hailin Zhang et al. “Experimental Analysis of Large-scale Learnable Vector Storage Compres- sion”. In: Proc. VLDB Endow. 17.4 (2023), pp. 808–

    2023

  165. [175]

    Large Language Model-Based Interpretable Machine Learning Con- trol in Building Energy Systems

    Liang Zhang and Zhelun Chen. “Large Language Model-Based Interpretable Machine Learning Con- trol in Building Energy Systems”. In:ArXiv abs/2402.09584 (2024)

    2024 arXiv

  166. [176]

    A survey on multimodal large language models

    Shukang Yin et al. “A survey on multimodal large language models”. In:arXiv preprint arXiv:2306.13549 (2023)

    2023 arXiv

  167. [177]

    PRISE: LLM-Style Sequence Compression for Learning Temporal Action Ab- stractions in Control

    Ruijie Zheng et al. “PRISE: LLM-Style Sequence Compression for Learning Temporal Action Ab- stractions in Control”. In: Forty-first Interna- tional Conference on Machine Learning

  168. [178]

    Revolutionizing Database Q&A with Large Language Models: Comprehen- sive Benchmark and Evaluation

    Yihang Zheng et al. “Revolutionizing Database Q&A with Large Language Models: Comprehen- sive Benchmark and Evaluation”. In:arXiv preprint arXiv:2409.04475 (2024)

    2024 arXiv

  169. [179]

    A comprehensive survey on pre- trained foundation models: A history from bert to chatgpt

    Ce Zhou et al. “A comprehensive survey on pre- trained foundation models: A history from bert to chatgpt”. In:arXiv preprint arXiv:2302.09419 (2023)

    2023 arXiv

  170. [180]

    Retrieval- augmented Query Reformulation for Heteroge- neous Research Asset Retrieval in Virtual Re- search Environment

    Peide Zhu, Na Li, and Zhiming Zhao. “Retrieval- augmented Query Reformulation for Heteroge- neous Research Asset Retrieval in Virtual Re- search Environment”. In: Companion Proceed- ings of the ACM on Web Conference 2024. 2024, pp. 907–910

    2024

  171. [181]

    Spine: Scaling up programming-by-negative-example for string filtering and transformation

    Chaoji Zuo, Sepehr Assadi, and Dong Deng. “Spine: Scaling up programming-by-negative-example for string filtering and transformation”. In: Proceed- ings of the 2022 International Conference on Man- agement of Data . 2022, pp. 521–530

    2022

  172. [182]

    Extract, De- fine, Canonicalize: An LLM-based Framework for Knowledge Graph Construction

    Bowen Zhang and Harold Soh. “Extract, De- fine, Canonicalize: An LLM-based Framework for Knowledge Graph Construction”. In: Pro- ceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, EMNLP 2024, Miami, FL, USA, November 12-16, 2024 . Ed. by Yaser...

    2024

  173. [186]

    ReAcTable: Enhancing Re- Act for Table Question Answering

    Yunjia Zhang et al. “ReAcTable: Enhancing Re- Act for Table Question Answering”. In: Proc. VLDB Endow. 17.8 (2024), pp. 1981–1994. doi: 10.14778/3659437.3659452

    2024

  174. [712]

    doi: 10.14778/3494124.3494149

  175. [822]

    doi: 10.14778/3636218.3636234

  176. [1145]

    doi: 10.14778/3641204.3641221

  177. [1764]

    doi: 10.14778/3654621.3654639

  178. [2023]

    OpenReview.net, 2023

    2023

  179. [2024]

    Revisiting Data Analysis with Pre-trained Foundation Models 25

    OpenReview.net, 2024. Revisiting Data Analysis with Pre-trained Foundation Models 25

    2024

  180. [2470]

    doi: 10.14778/3603581.3603587

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.