StreamLink: Large-Language-Model Driven Distributed Data Engineering System

Dawei Feng; Di Mei; Huiri Tan; Lei Ren; Xianying Lou; Zhangxi Tan

REVIEW 5 major objections 5 minor 31 references

StreamLink: Large-Language-Model Driven Distributed Data Engineering System

T0 review · 5 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

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

Pith's one-line read A locally hosted 8-billion-parameter LLM reaches 86.9% exact match and 89.7% execution accuracy on the Spider dev set, the best results in the paper's comparison table.

desk verdict Plausible system, unverifiable headline numbers: the Spider result may be real but the paper does not provide enough to check. read the letter →

arxiv 2505.21575 v1 pith:LLCI6MD7 submitted 2025-05-27 cs.DB cs.AI

Dawei Feng , Di Mei , Huiri Tan , Lei Ren , Xianying Lou , Zhangxi Tan This is my paper · ORCID

classification cs.DBcs.AI

keywords DistributedDatabaseLargeLanguageModelSQLGenerationLLM-DrivenCheckerNaturaltoLoRAFine-TuningPatentDataRetrievalPrivacy

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

StreamLink is a data system that lets users query very large databases by typing ordinary English instead of writing SQL. The paper's central claim is that a small language model fine-tuned locally, rather than a cloud AI service, can generate accurate SQL from natural language: on the dev set of the standard Spider text-to-SQL benchmark, its best model, SSQLG3.1-8B, scores 86.9% exact match and 89.7% execution accuracy, surpassing all listed baselines. The same system, deployed over 180 million patent records across three storage nodes, returns a user's requested patent in an average of under six seconds. A Llama-based syntax and security checker filters generated SQL to block malformed or malicious statements before execution. If true, this makes large private databases directly queryable by non-programmers without sending data to an outside service.

What carries the argument

The central mechanism is the fine-tuned local language model used as an NL-to-SQL generator (SSQLG), trained with LoRA on a 1:1 hybrid of domain-augmented query templates and public datasets such as WikiSQL and Spider. The generator is paired with a Llama-based SQL syntax and security checker (SSQLC) that validates each statement before execution. Around these models, StreamLink layers a distributed stack, with Spark for computing, HDFS-backed storage, a load-balanced WebUI cluster, and a Central Control Unit, so that a natural-language request becomes SQL, is checked, and runs over 180 million patents quickly.

What would settle it

Run the published model, if one is released, on the Spider dev set with the standard exact-match and execution-accuracy evaluation and compare the results to the reported 86.9% and 89.7%; also inspect the fine-tuning mixture for any overlap with Spider dev databases or questions.

Watch

Extended reading notes

Core claim

On the paper's own terms, the discovery is that a modestly sized, locally hosted LLM can be adapted to structured-data querying well enough to beat specialized text-to-SQL systems on a standard benchmark. The best system, SSQLG3.1-8B, fine-tuned from Llama-3.1-8B, reaches 86.9% exact match (verbatim agreement with a gold SQL query) and 89.7% execution accuracy (the query returns the right result) on the Spider dev set, the best numbers in Table 1. The same deployment, running on three storage nodes totaling 15.3 TB with 280 cores, answers natural-language queries over roughly 180 million patents in an average of under six seconds, and its Llama-3-based checker, SSQLC3-8B, catches 98.09% of malicious SQL statements at 79.31% precision. The paper frames these results as evidence that private-by-design, domain-adapted LLM assistants can make large distributed databases accessible to non-specialists without sacrificing security or scale.

Load-bearing premise

The central claim holds only if the reported Spider scores come from a fair, leak-free comparison, meaning the same evaluation setup for every model and no Spider dev-set example in the training data, and the paper gives no code, model, or evaluation script to verify that.

Editorial extensions

If this is right

Users who do not know SQL can retrieve specific records from huge private databases using natural language, removing a major barrier to data-driven work.
Keeping the LLM on local infrastructure avoids sending proprietary or sensitive data to a public AI service, which is the paper's stated privacy design.
The two-part generation-plus-checking pipeline makes LLM-produced SQL safe enough to run automatically, since malformed or injection-style statements are filtered before execution.
Because the system continues to use Spark and HDFS underneath, the approach inherits distributed scalability and can be pointed at other large datasets with new schema templates.
Domain adaptation can be rapid: given a small set of query templates, the bi-directional propagation method grows a training set and fine-tunes a model for that database.

Reading between the lines

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

If the reported accuracy holds up, the 1:1 hybrid ratio is a transferable recipe: a domain with only a few dozen query templates could be made queryable by generating a synthetic parallel corpus instead of hiring SQL experts.
The checker results suggest a practical follow-up: since SSQLC3-8B has 79% precision, a production system would want a low-cost second-pass filter or a human-in-the-loop confirmation for queries the checker flags, otherwise roughly one in five legitimate queries may be blocked.
A natural extension beyond patents is to apply the same template-propagation fine-tuning to other governed data, such as clinical records or financial reporting, where the privacy argument for local LLMs is even stronger.
The end-to-end six-second latency mixes LLM generation, checking, and Spark execution; splitting those stages in a benchmark would show whether the bottleneck is model inference or data scan, and where a faster deployment should invest.

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.

Desk Editor's Note

read the letter

Quick read for you. StreamLink is a workmanlike integration of local LLMs (Llama-3.1-8B fine-tuned with LoRA) with Spark/Hadoop for NL-to-SQL over a large patent warehouse. The architecture is sensible: a CCU schedules tasks across LLM and Spark clusters, a Nginx WebUI cluster balances load, and a Llama-based syntax/security checker sits in front of execution. Deploying this over 180 million patents and reporting sub-6-second average latency is a real engineering achievement, and the privacy rationale for local models is sound. Credit where due: the template-augmentation for domain adaptation is a practical trick, and the 1:1 hybrid ratio with open-domain sets is a reasonable choice.

The soft spots are where the claims outrun the evidence. The headline result—86.9% EM / 89.7% EA on Spider dev—is reported in Table 1 with no code, no model weights, no hyperparameters, no evaluation harness, and no error bars. More worrying, Section 3.1 says the fine-tuning corpus combined domain data with 'WikiSQL and Spider' but never states that 'Spider' means the official training split. If dev-set schemas or questions leaked into the augmentation, the over-10% claim collapses. The paper doesn't have to prove a negative, but it has to specify the split; it doesn't. The baselines are also dated (T5-3B, PICARD, etc.), so an apples-to-apples comparison with a current 8B model is missing. And the 'over 10%' phrasing is ambiguous—percentage points vs. relative improvement.

The security checker section is internally coherent but compares only its own variants; no existing SQL injection detector. The Kaggle dataset is fine for a smoke test, not for a claim about real-world security. Several typos (duplicated FN definition, inconsistent equation fonts) are minor.

Bottom line: the system story is credible, the benchmark story is not checkable. This deserves a serious referee only if the authors commit to releasing the training/evaluation artifacts and explicitly state the Spider split. As a desk decision, I'd send it out with that condition, expecting heavy revision. If the authors won't release anything, reject. For you personally: maybe worth a quick skim for the system integration, but don't cite the Spider numbers yet.

Referee Report

5 major / 5 minor

Summary. StreamLink is presented as an LLM-driven distributed data engineering system built on Spark and Hadoop, with a WebUI cluster, a Central Control Unit, and a locally deployed fine-tuned LLM (Llama-3.1-8B via LoRA) that translates natural-language requests into SQL. The paper reports 86.9% exact match and 89.7% execution accuracy on the Spider dev set, claims improvements of over 10% over baselines, describes zero-shot Llama-based SQL syntax and security checkers evaluated on a Kaggle SQL-injection dataset, and claims end-to-end retrieval from 180 million patents in under 6 seconds.

Significance. The practical scenario is relevant and the local-LLM privacy design is a sensible contribution, as is the use of standard public benchmarks (Spider dev, Kaggle SQL injection) for falsifiable headline numbers. The paper's strengths are its deployment scale (180 million patents, 15.3 TB, 280 cores), the template-based domain augmentation idea, and the zero-shot security-checker evaluation. Those strengths do not yet establish the central claims, because no code, model weights, prompts, training corpus, or evaluation harness are provided; the Spider fine-tuning split is unspecified, so dev-set leakage cannot be ruled out; and the latency claim is presented without a measurement protocol.

major comments (5)

[§3.1, §4.1, Table 1] The training corpus for SSQLG3.1-8B is not specified at the level needed to validate the headline Spider result. Section 3.1 says only that the domain-specific dataset was combined with publicly available NL-2-SQL datasets like WikiSQL and Spider, and Section 4.1 evaluates on the Spider dev set without stating whether Spider means the official training split or the full dataset. Since the dev set is public, any use of dev questions, dev schemas, or templates derived from dev schemas during fine-tuning would contaminate the 86.9% EM and 89.7% EA numbers, and the absence of released code, model weights, training corpus, and evaluation scripts makes this impossible to check. The paper must specify the exact split, describe the template-augmented data, support the claimed optimal 1:1 hybrid ratio with the missing ablation, and either release artifacts or provide an independent reproduction before the central claim can be accepted.
[§4.1, Table 1] The comparison is not apples-to-apples. The baselines (T5-3B, BERT_LARGE+RAT-SQL, PICARD, RASAT, and others) are from earlier model families and were evaluated under their own pipelines, while SSQLG3.1-8B is a LoRA-adapted Llama-3.1-8B with undisclosed hyperparameters (LoRA rank, alpha, learning rate, epochs, batch size, prompt format, and decoding settings). The text claims the model exceeded the baseline by over 10% in both execution accuracy and exact match, but Table 1 shows an absolute execution-accuracy lead of 9.2 percentage points over the best baseline (89.7 vs. 80.5); the claim is only true under a relative-improvement interpretation. Please report controlled experiments in a single harness, include contemporary LLM baselines, and provide variance estimates or error bars.
[§4.2, Table 2, Figures 4 and 5] The malicious-SQL evaluation is reported without a reproducible protocol: the zero-shot prompts, classification threshold, sampling scheme, and number of repetitions are not given, and it is unclear whether the syntax checker and security checker were evaluated jointly or separately. All metrics in Table 2 are point estimates without confidence intervals, so the deployment recommendation of SSQLC3-8B is not statistically supported. The prose interpretation that Llama-3.1's training data is not as well-optimized for SQL interception is unsupported speculation and should be removed or replaced with evidence.
[§1, §3.2] The headline system-level claim that natural-language requests retrieve items from 180 million patents in under 6 seconds is not supported by any experimental protocol. The paper does not define the query workload, the number of runs, the hardware configuration used for the measurement, the latency percentiles, or the comparison point, so the claim cannot be verified or reproduced. A system paper that advertises this as a contribution needs at least a defined benchmark, a latency distribution, and a description of what is included in the measured interval.
[§1, §3.2, Figure 1] The Central Control Unit (CCU), listed as a contribution, is never specified. The text states that it schedules tasks across the distributed LLM, UI, and storage clusters, but gives no information about its implementation, scheduling policy, interfaces, or failure behavior, and Figure 1 only shows it as a box. Without this, the claimed architectural contribution cannot be assessed.

minor comments (5)

[§3.1, Equations (1) and (2)] The word 'correspoding' should be 'corresponding', and the probability notation should be unified between the two equations (lowercase p in Equation (1) versus uppercase P in Equation (2)).
[§4.2, Equations (3)-(6)] In the metric definitions, the second 'FN (False Negative)' bullet is repeated; the intended entry appears to be 'TN (True Negative) - Negative in the label, and predicted negative'.
[Table 1] The column header 'Exact Match Accuracy' is ambiguous because the table has two numeric columns; label them explicitly as 'Exact Match' and 'Execution Accuracy'.
[Figure 4] The caption calls the panels 'Precision-Recall-Escape-Misintercept Curves', but each panel plots a metric against the number of tests; either change the plot type to standard curves or rename the figure.
[References] References [23] and [24] are not cited in the text, and no reference is provided for Llama-3 or Llama-3.1, which are the bases for the main results.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the central benchmark claims are measured against external standard datasets and are not derived from fitted inputs.

full rationale

The paper's main quantitative claims are evaluated against external, standard benchmarks: the Spider dev-set (Section 4.1, Table 1) and the Kaggle SQL injection dataset (Section 4.2, Table 2). These benchmarks are not defined in terms of the model's own outputs or fitted parameters, so reporting exact match and execution accuracy on them is an independent evaluation. The fine-tuning corpus mixes a domain-specific patent dataset with WikiSQL and Spider, and the 1:1 hybrid ratio is described as an experimentally chosen internal setting; there is no equation or construction showing that the reported 86.9% EM / 89.7% EA numbers are forced by this ratio. The paper does not rely on self-citations or imported uniqueness theorems to justify its design choices. The '<6 seconds' system latency claim is a deployment measurement on a private 180-million-patent database, not a consequence of any fitted parameter. While the manuscript leaves reproducibility details unclear (e.g., whether 'Spider' means the official training split, and no released artifacts), that is a verifiability concern, not circularity. No step was found where a claim reduces by construction to its own input.

Assumptions & free parameters 2 free parameters · 3 assumptions · 1 invented entities

The central claims rest on external benchmarks (Spider and Kaggle) plus a set of undisclosed training choices. The 1:1 hybrid ratio and LoRA hyperparameters are free parameters that can affect the headline accuracy. The paper also assumes that generic benchmark performance transfers to the patent domain and that public SQL injection data represent real-world attacks, neither of which is tested.

free parameters (2)

Hybrid training ratio (domain-specific to open-domain datasets) = 1:1
Reported in Section 3.1 as the optimal ratio after "extensive experiments", but the experiments and variance are not shown; this choice affects the fine-tuned model's accuracy.
LoRA fine-tuning hyperparameters (rank, alpha, learning rate, epochs, batch size)
Not reported anywhere; the headline Spider accuracy of SSQLG3.1-8B depends on these choices, so they are free parameters from the reader's perspective.

assumptions (3)

domain assumption Spider dev-set accuracy is a valid proxy for the system's performance on the real patent domain.
The system is deployed on patent data, but the fine-tuned models are only evaluated on the generic Spider benchmark; no patent-specific NL-to-SQL evaluation is reported.
domain assumption The Kaggle SQL injection dataset is representative of real-world malicious SQL traffic.
Section 4.2 uses this public dataset for zero-shot evaluation and generalizes to deployment security without testing on actual production traffic.
domain assumption Standard distributed-computing assumptions for Apache Spark/Hadoop hold in this deployment.
The paper assumes the Spark/Hadoop stack provides the claimed scalability and fault tolerance, but no load test or failure injection is reported.

invented entities (1)

Central Control Unit (CCU)
purpose: Schedules and coordinates tasks among the WebUI cluster, LLM cluster, and Spark computing cluster (Section 1, Figure 1).
Introduced as a new architectural component, but no experiment isolates its contribution or measures its overhead; it is a system design element rather than a falsifiable entity.

how reviews work

0 comments

Cite this review

Pith. "Pith review of StreamLink: Large-Language-Model Driven Distributed Data Engineering System." pith.science (2026). https://pith.science/paper/LLCI6MD7

@misc{pith2026250521575,
  author       = {Pith},
  title        = {Pith review of: StreamLink: Large-Language-Model Driven Distributed Data Engineering System},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/LLCI6MD7}},
  note         = {Machine review of arXiv:2505.21575}
}

read the original abstract

Large Language Models (LLMs) have shown remarkable proficiency in natural language understanding (NLU), opening doors for innovative applications. We introduce StreamLink - an LLM-driven distributed data system designed to improve the efficiency and accessibility of data engineering tasks. We build StreamLink on top of distributed frameworks such as Apache Spark and Hadoop to handle large data at scale. One of the important design philosophies of StreamLink is to respect user data privacy by utilizing local fine-tuned LLMs instead of a public AI service like ChatGPT. With help from domain-adapted LLMs, we can improve our system's understanding of natural language queries from users in various scenarios and simplify the procedure of generating database queries like the Structured Query Language (SQL) for information processing. We also incorporate LLM-based syntax and security checkers to guarantee the reliability and safety of each generated query. StreamLink illustrates the potential of merging generative LLMs with distributed data processing for comprehensive and user-centric data engineering. With this architecture, we allow users to interact with complex database systems at different scales in a user-friendly and security-ensured manner, where the SQL generation reaches over 10\% of execution accuracy compared to baseline methods, and allow users to find the most concerned item from hundreds of millions of items within a few seconds using natural language.

Figures

Figures reproduced from arXiv: 2505.21575 by the authors.

**Figure 1.** Architecture of Project StreamLink redundancy for reliability), and the system is accelerated using 280 cores (560 threads) and 2.6TB of memory. During testing, we confirmed that the average time from user input in natural language to obtaining the desired patent from a database containing over 180 million patents is within 6 seconds. • Data Privacy Protecting: We have confirmed that building a localized LLM-based … view at source ↗

**Figure 2.** Data augmentation via bi-directional propagation [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

**Figure 3.** Distrubuted system architecture with LLM to im [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗

Figures from the paper (2 more)

**Figure 5.** Figure 5: ROC Curve of our SSQLC methods, and the AUC of [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]

**Figure 4.** Figure 4: Malicious SQL interception analyzing on our LLM [PITH_FULL_IMAGE:figures/full_fig_p006_4.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

31 extracted references · 20 canonical work pages

[1]

Language models are few-shot learners.Advances in neural infor- mation processing systems, 33:1877–1901, 2020

Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners.Advances in neural infor- mation processing systems, 33:1877–1901, 2020

1901
[2]

Spark: Cluster computing with working sets

Matei Zaharia, Mosharaf Chowdhury, Michael J Franklin, Scott Shenker, and Ion Stoica. Spark: Cluster computing with working sets. In2nd USENIX Workshop on Hot Topics in Cloud Computing (HotCloud 10), 2010

work page 2010
[3]

Data-intensive applications, challenges, techniques and technologies: A survey on big data.Information sciences, 275:314– 347, 2014

CL Philip Chen and Chun-Yang Zhang. Data-intensive applications, challenges, techniques and technologies: A survey on big data.Information sciences, 275:314– 347, 2014

work page 2014
[4]

Data-intensive science: a new paradigm for biodiversity studies.BioScience, 59(7):613–620, 2009

Steve Kelling, Wesley M Hochachka, Daniel Fink, Mirek Riedewald, Rich Caruana, Grant Ballard, and Giles Hooker. Data-intensive science: a new paradigm for biodiversity studies.BioScience, 59(7):613–620, 2009

work page 2009
[5]

Oxford University Press, 2007

Arthur Stone, Saul Shiffman, Audie Atienza, and Linda Nebeling.The science of real-time data capture: Self-reports in health research. Oxford University Press, 2007

work page 2007
[6]

Big data storage technologies: a survey.Frontiers of Information Technology & Electronic Engineering, 18:1040– 1070, 2017

Aisha Siddiqa, Ahmad Karim, and Abdullah Gani. Big data storage technologies: a survey.Frontiers of Information Technology & Electronic Engineering, 18:1040– 1070, 2017

work page 2017
[7]

Big data visualization: Tools and challenges

Syed Mohd Ali, Noopur Gupta, Gopal Krishna Nayak, and Rakesh Kumar Lenka. Big data visualization: Tools and challenges. In2016 2nd International conference on contemporary computing and informatics (IC3I), pages 656–660. IEEE, 2016

work page 2016
[8]

The uspto patent assignment dataset: Descriptions and analysis

Alan C Marco, Amanda Myers, Stuart JH Graham, Paul D’Agostino, and Kirsten Apple. The uspto patent assignment dataset: Descriptions and analysis. 2015

work page 2015

Show all 31 references

[9]

Structured query language (sql) fundamentals.Current protocols in bioinformatics, (1):9–2, 2003

D Curtis Jamison. Structured query language (sql) fundamentals.Current protocols in bioinformatics, (1):9–2, 2003

2003
[10]

Graphix-t5: Mixing pre-trained transformers with graph-aware layers for text-to-sql parsing

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

2023
[11]

N-best hypotheses reranking for text-to-sql systems

Lu Zeng, Sree Hari Krishnan Parthasarathi, and Dilek Hakkani-Tur. N-best hypotheses reranking for text-to-sql systems. In2022 IEEE Spoken Language Technology Workshop (SLT), pages 663–670. IEEE, 2023

2023
[12]

Importance of synthe- sizing high-quality data for text-to-sql parsing.arXiv preprint arXiv:2212.08785, 2022

Yiyun Zhao, Jiarong Jiang, Yiqun Hu, Wuwei Lan, Henry Zhu, Anuj Chauhan, Alexander Li, Lin Pan, Jun Wang, Chung-Wei Hang, et al. Importance of synthe- sizing high-quality data for text-to-sql parsing.arXiv preprint arXiv:2212.08785, 2022

2022 arXiv
[13]

Llama 2: Open foundation and fine-tuned chat models.arXiv preprint arXiv:2307.09288, 2023

Hugo Touvron, Louis Martin, Kevin Stone, Peter Albert, Amjad Almahairi, Yas- mine Babaei, Nikolay Bashlykov, Soumya Batra, Prajjwal Bhargava, Shruti Bhos- ale, et al. Llama 2: Open foundation and fine-tuned chat models.arXiv preprint arXiv:2307.09288, 2023

2023 arXiv
[14]

Sentence-bert: Sentence embeddings using siamese bert-networks.arXiv preprint arXiv:1908.10084, 2019

Nils Reimers and Iryna Gurevych. Sentence-bert: Sentence embeddings using siamese bert-networks.arXiv preprint arXiv:1908.10084, 2019

1908 arXiv
[15]

Layout evaluation of large capacity warehouses.Facilities, 25(7/8):259–270, 2007

José Ignacio Huertas, Jenny Díaz Ramírez, and Federico Trigos Salazar. Layout evaluation of large capacity warehouses.Facilities, 25(7/8):259–270, 2007

2007
[16]

Sql: From traditional databases to big data

Yasin N Silva, Isadora Almeida, and Michell Queiroz. Sql: From traditional databases to big data. InProceedings of the 47th ACM Technical Symposium on Computing Science Education, pages 413–418, 2016

2016
[17]

Apache Software Foundation. Hadoop
[18]

Valuenet: A natural language-to-sql system that learns from database information

Ursin Brunner and Kurt Stockinger. Valuenet: A natural language-to-sql system that learns from database information. In2021 IEEE 37th International Conference on Data Engineering (ICDE), pages 2177–2182. IEEE, 2021

2021
[19]

Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021

Edward J Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. Lora: Low-rank adaptation of large language models.arXiv preprint arXiv:2106.09685, 2021

2021 arXiv
[20]

Seq2sql: Generating struc- tured queries from natural language using reinforcement learning.CoRR, abs/1709.00103, 2017

Victor Zhong, Caiming Xiong, and Richard Socher. Seq2sql: Generating struc- tured queries from natural language using reinforcement learning.CoRR, abs/1709.00103, 2017

2017 arXiv
[21]

Spider: A large-scale human- labeled dataset for complex and cross-domain semantic parsing and text-to-sql task.arXiv preprint arXiv:1809.08887, 2018

Tao Yu, Rui Zhang, Kai Yang, Michihiro Yasunaga, Dongxu Wang, Zifan Li, James Ma, Irene Li, Qingning Yao, Shanelle Roman, et al. Spider: A large-scale human- labeled dataset for complex and cross-domain semantic parsing and text-to-sql task.arXiv preprint arXiv:1809.08887, 2018

2018 arXiv
[22]

Nginx: the high-performance web server and reverse proxy.Linux Journal, 2008(173):2, 2008

Will Reese. Nginx: the high-performance web server and reverse proxy.Linux Journal, 2008(173):2, 2008

2008
[23]

Round robin scheduling–a survey

Rasmus V Rasmussen and Michael A Trick. Round robin scheduling–a survey. European Journal of Operational Research, 188(3):617–636, 2008

2008
[24]

Efficient and scalable multiprocessor fair scheduling using distributed weighted round-robin.ACM Sigplan Notices, 44(4):65–74, 2009

Tong Li, Dan Baumberger, and Scott Hahn. Efficient and scalable multiprocessor fair scheduling using distributed weighted round-robin.ACM Sigplan Notices, 44(4):65–74, 2009

2009
[25]

Natural sql: Making sql easier to infer from natural language specifications.arXiv preprint arXiv:2109.05153, 2021

Yujian Gan, Xinyun Chen, Jinxia Xie, Matthew Purver, John R Woodward, John Drake, and Qiaofu Zhang. Natural sql: Making sql easier to infer from natural language specifications.arXiv preprint arXiv:2109.05153, 2021

2021 arXiv
[26]

Grappa: Grammar-augmented pre-training for table semantic parsing.arXiv preprint arXiv:2009.13845, 2020

Tao Yu, Chien-Sheng Wu, Xi Victoria Lin, Bailin Wang, Yi Chern Tan, Xinyi Yang, Dragomir Radev, Richard Socher, and Caiming Xiong. Grappa: Grammar-augmented pre-training for table semantic parsing.arXiv preprint arXiv:2009.13845, 2020

2009 arXiv
[27]

𝑠2sql: Injecting syntax to question-schema interaction graph encoder for text-to-sql parsers.arXiv preprint arXiv:2203.06958, 2022

Binyuan Hui, Ruiying Geng, Lihan Wang, Bowen Qin, Bowen Li, Jian Sun, and Yongbin Li. 𝑠2sql: Injecting syntax to question-schema interaction graph encoder for text-to-sql parsers.arXiv preprint arXiv:2203.06958, 2022

2022 arXiv
[28]

Picard: Parsing incrementally for constrained auto-regressive decoding from language models

Torsten Scholak, Nathan Schucher, and Dzmitry Bahdanau. Picard: Parsing incrementally for constrained auto-regressive decoding from language models. arXiv preprint arXiv:2109.05093, 2021

2021 arXiv
[29]

Rasat: Integrating rela- tional structures into pretrained seq2seq model for text-to-sql.arXiv preprint arXiv:2205.06983, 2022

Jiexing Qi, Jingyao Tang, Ziwei He, Xiangpeng Wan, Yu Cheng, Chenghu Zhou, Xinbing Wang, Quanshi Zhang, and Zhouhan Lin. Rasat: Integrating rela- tional structures into pretrained seq2seq model for text-to-sql.arXiv preprint arXiv:2205.06983, 2022

2022 arXiv
[30]

Structure-grounded pretraining for text-to- sql.arXiv preprint arXiv:2010.12773, 2020

Xiang Deng, Ahmed Hassan Awadallah, Christopher Meek, Oleksandr Polozov, Huan Sun, and Matthew Richardson. Structure-grounded pretraining for text-to- sql.arXiv preprint arXiv:2010.12773, 2020

2010 arXiv
[31]

Bert: Pre- training of deep bidirectional transformers for language understanding.arXiv preprint arXiv:1810.04805, 2018

Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. Bert: Pre- training of deep bidirectional transformers for language understanding.arXiv preprint arXiv:1810.04805, 2018

2018 arXiv

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.

[1] [1]

Language models are few-shot learners.Advances in neural infor- mation processing systems, 33:1877–1901, 2020

Tom Brown, Benjamin Mann, Nick Ryder, Melanie Subbiah, Jared D Kaplan, Prafulla Dhariwal, Arvind Neelakantan, Pranav Shyam, Girish Sastry, Amanda Askell, et al. Language models are few-shot learners.Advances in neural infor- mation processing systems, 33:1877–1901, 2020

1901

[2] [2]

Spark: Cluster computing with working sets

Matei Zaharia, Mosharaf Chowdhury, Michael J Franklin, Scott Shenker, and Ion Stoica. Spark: Cluster computing with working sets. In2nd USENIX Workshop on Hot Topics in Cloud Computing (HotCloud 10), 2010

work page 2010

[3] [3]

Data-intensive applications, challenges, techniques and technologies: A survey on big data.Information sciences, 275:314– 347, 2014

CL Philip Chen and Chun-Yang Zhang. Data-intensive applications, challenges, techniques and technologies: A survey on big data.Information sciences, 275:314– 347, 2014

work page 2014

[4] [4]

Data-intensive science: a new paradigm for biodiversity studies.BioScience, 59(7):613–620, 2009

Steve Kelling, Wesley M Hochachka, Daniel Fink, Mirek Riedewald, Rich Caruana, Grant Ballard, and Giles Hooker. Data-intensive science: a new paradigm for biodiversity studies.BioScience, 59(7):613–620, 2009

work page 2009

[5] [5]

Oxford University Press, 2007

Arthur Stone, Saul Shiffman, Audie Atienza, and Linda Nebeling.The science of real-time data capture: Self-reports in health research. Oxford University Press, 2007

work page 2007

[6] [6]

Big data storage technologies: a survey.Frontiers of Information Technology & Electronic Engineering, 18:1040– 1070, 2017

Aisha Siddiqa, Ahmad Karim, and Abdullah Gani. Big data storage technologies: a survey.Frontiers of Information Technology & Electronic Engineering, 18:1040– 1070, 2017

work page 2017

[7] [7]

Big data visualization: Tools and challenges

Syed Mohd Ali, Noopur Gupta, Gopal Krishna Nayak, and Rakesh Kumar Lenka. Big data visualization: Tools and challenges. In2016 2nd International conference on contemporary computing and informatics (IC3I), pages 656–660. IEEE, 2016

work page 2016

[8] [8]

The uspto patent assignment dataset: Descriptions and analysis

Alan C Marco, Amanda Myers, Stuart JH Graham, Paul D’Agostino, and Kirsten Apple. The uspto patent assignment dataset: Descriptions and analysis. 2015

work page 2015