Guixu: Valuation-Driven Data Discovery for Autonomous AI Agents with On-Chain Attestation

Jiaqi Chai; Ke Chen; Lidan Shou; Xilin Li; Yifan Wu; Yuchen Peng; Yufei Qian

REVIEW 3 major objections 4 minor 70 references

Guixu: Valuation-Driven Data Discovery for Autonomous AI Agents with On-Chain Attestation

T0 review · 3 major / 4 minor · reviewed 2026-08-12 · deepseek-v4-flash

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

Pith's one-line read Guixu claims that data discovery for autonomous agents should be a valuation problem: score each dataset for task utility, choose a portfolio under budget, and verify the purchase on-chain.

desk verdict Cleanly written demo integration of known valuation and procurement pieces; the accuracy claims are asserted, not measured, so the paper needs either data or softer claims before it can be taken at face value. read the letter →

arxiv 2608.07949 v1 pith:D5M4T7LT submitted 2026-08-08 cs.AI cs.CRcs.DBcs.IRcs.MA

Yifan Wu , Yuchen Peng , Jiaqi Chai , Yufei Qian , Xilin Li , Ke Chen , Lidan Shou This is my paper · ORCID

classification cs.AIcs.CRcs.DBcs.IRcs.MA

keywords datavaluationautonomousagentsdatasetdiscoverybudget-constrainedoptimizationknapsackproxy-labelpropagationon-chainattestationagenticpayment

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

Guixu argues that dataset discovery for autonomous agents should be reframed from a retrieval problem into a task-aware valuation and procurement problem. The paper presents a working system that takes a natural-language task, searches heterogeneous data sources, scores each candidate dataset for task-specific utility, and selects a budget-constrained portfolio via knapsack optimization. It claims this valuation-driven selection yields measurable accuracy gains over keyword-based retrieval even at zero budget, and that the full purchase loop can run autonomously with on-chain attestation for trust. If the system works as described, agents could independently acquire verified training data for niche tasks end to end.

What carries the argument

The three-phase valuation pipeline is the load-bearing mechanism. Phase 1 scores candidates by weighted signals (relevance, schema fit, data scale, label quality, metadata completeness, on-chain reputation) with Shapley-valued weights. Phase 2 propagates LLM-oracle scores from high-score and low-score anchors to similar unscored samples when the similarity gap exceeds a threshold, reserving oracle calls for ambiguous samples. Phase 3 solves a two-dimensional knapsack over price and size via dynamic programming, advancing to the next coarse-ranked batch if no feasible portfolio exists. On-chain attestation supplies the reputation signal that feeds back into Phase 1.

What would settle it

Run the LLM oracle on every sample of a candidate dataset and compare those scores with Guixu's propagated scores: if mean absolute error or ranking disagreement is large, the proxy-label phase is not preserving fidelity and the measured accuracy gain cannot be attributed to valuation. A second check: repeat the zero-budget scenario with random dataset selection and with keyword search; if accuracy is equal, the valuation pipeline itself adds nothing beyond retrieval.

Watch

Extended reading notes

Core claim

The central claim is that valuation-driven data discovery outperforms keyword-based retrieval because it optimizes what actually matters for the downstream task: per-sample utility, budget, size constraints, and trust. The paper's demonstrator takes a natural-language task such as 'train a safety-helmet classifier with a $2.00 budget,' parses it into task and data schemas, searches sources like Kaggle and Hugging Face, scores each candidate through three phases (coarse ranking on six signals, proxy-label propagation with an LLM oracle for ambiguous samples, and multi-round two-dimensional knapsack optimization), and then executes the purchase through escrow smart contracts, Shamir key sharing, and on-chain attestation that feeds back into future valuations. It claims this yields a measurable accuracy gain even at zero budget and enables agents to move beyond keyword-based retrieval toward task- and budget-aware, trustworthy procurement.

Load-bearing premise

The cost-saving argument rests on propagated proxy scores being close to what the LLM oracle would assign, yet the paper provides no measurement of propagation accuracy or threshold calibration.

Editorial extensions

If this is right

An agent can complete a niche task end to end, from a natural-language request to a trained model, without human selection of data.
Buyers can make purchase decisions that jointly optimize task utility, price, and trust, instead of picking datasets by keyword hits.
Proxy-label propagation makes per-sample valuation affordable enough for realistic datasets, because only ambiguous samples trigger expensive oracle calls.
The multi-round knapsack handles infeasible budgets by expanding the candidate pool until a feasible portfolio is found.
On-chain attestation creates a feedback loop where past purchases improve future valuation decisions.

Reading between the lines

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

The valuation pipeline is not tied to image classification; the same schema-based scoring and knapsack selection could be applied to tabular or text datasets, though the paper only demonstrates vision tasks.
If on-chain reputation feeds valuation, sellers with good attestations may command price premiums, a market dynamic the paper does not explore.
A direct measurement of propagated versus oracle scores would let operators tune the threshold as an explicit cost-accuracy knob, a calibration study the paper leaves implicit.
The zero-budget accuracy claim would be stronger if compared against a human-curated baseline or a simple budget-agnostic ranker; the paper reports the result without that ablation.

Share X Bluesky LinkedIn Reddit HN

Signed reviews

No signed human review yet.

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

The paper is a system demonstration, and that is the right lens to read it through. The genuinely new thing is the integration: a three-phase valuation pipeline (coarse ranking with Shapley-weighted signals, proxy-label propagation for cheap scoring, multi-round knapsack for budget-constrained selection) wrapped in an MCP server with agentic payments and on-chain escrow, all feeding trust signals back into the valuation. Each component is known, but nobody has put them together in one agent-facing data discovery workflow. The writing is clear, the architecture is concrete, and the authors are honest about the prior work they build on. The GitHub artifact link is a plus, though the paper does not tell us what is actually in it. The soft spots are real but proportionate to a demo paper. The big one is Section 3, where the zero-budget scenario claims attendees can observe that valuation-driven selection yields a measurable accuracy gain. That is an empirical claim with no number, no baseline, no test set, and no repeat protocol. A live demo is not a measurement; any perceived gain could come from dataset ordering, the Shapley weights, or the LLM oracle's seed scoring rather than the propagation and knapsack layers that are the contribution. The same goes for Phase 2: the paper asserts that proxy-label propagation yields substantial cost savings without sacrificing scoring fidelity, but never measures agreement between propagated and oracle scores, and never describes how the threshold theta is calibrated. The Shapley weights are derived from 'historical dataset performance,' which is also left unspecified. These are load-bearing for the paper's value proposition, so they matter, but for a demo track they are fixable: either report a small-scale evaluation with baselines, or tone down the claims to what the demo actually shows. The paper is not internally inconsistent, and the math that is described (two-dimensional knapsack via DP) is standard and correct. My main disagreement with a strict reading would be to note that in the demo-paper genre, the absence of experiments is not a fatal flaw unless the authors pretend otherwise. Here they do pretend, in one sentence, so the fix is easy. Who is this for? Readers working on agentic data management, data valuation, or blockchain-based data markets. It is a plausible reference point for future systems, and it deserves a serious referee. My recommendation: send it to peer review, and ask the authors to either supply a minimal empirical check (even a handful of datasets with baseline comparisons) or rewrite the zero-budget claim as a demonstration of workflow rather than a measured accuracy gain.

Referee Report

3 major / 4 minor

Summary. This paper presents Guixu, a demonstration system for valuation-driven dataset discovery for autonomous AI agents. The system combines a model-context-protocol interface, a three-phase valuation pipeline (multi-signal coarse ranking, proxy-label propagation, and multi-round knapsack optimization), and an on-chain escrow/attestation layer. The authors argue that this architecture moves beyond keyword-based dataset retrieval toward task-aware, budget-constrained, and trust-aware data procurement. The paper describes the architecture, the valuation formulas, the knapsack optimization, and two interactive demonstration scenarios, and it claims that valuation-driven selection yields a measurable accuracy gain even at zero budget.

Significance. If the claims are substantiated, Guixu addresses a genuine gap: existing data discovery platforms are retrieval-oriented and do not support task-aware utility estimation, budget-constrained selection, or trust-aware procurement. The proposed pipeline is a plausible integration of known techniques (embedding similarity, proxy labels, DP-knapsack, Shapley valuation, and blockchain escrow) into an agent-usable workflow. The paper also ships an artifact link and describes concrete system components (MCP server, x402/MPP payment routing, Shamir-secret-sharing key distribution). However, the core novel claims—that proxy-label propagation preserves scoring fidelity and that the overall selection yields measurable accuracy gains—are not backed by experimental evidence in the manuscript. As a demonstration paper, the system architecture is interesting, but the quantitative value proposition remains unverified.

major comments (3)

[Section 3, Scenario 1] The central claim that "the valuation-driven selection yields a measurable accuracy gain even at zero budget" is asserted but never measured. No baseline, test dataset, accuracy metric, or repeated trials are reported anywhere in the paper. Since this claim is the main motivation for the entire valuation pipeline, the demonstration section must either include a concrete experimental protocol with results (e.g., comparing against a keyword-retrieval baseline on a standard classification task) or explicitly relabel this as an anecdotal observation rather than a measurable gain.
[Section 2.2, Phase 2] The proxy-label propagation stage claims "substantial cost savings without sacrificing scoring fidelity," but no evidence is given that propagated scores agree with oracle scores. There is no measurement of propagation accuracy, no calibration of the threshold θ, and no validation that the textual-similarity gap Δ correlates with task utility. Because propagated scores directly determine the per-dataset quality value q̂_i that feeds the knapsack optimization, this omission is load-bearing. The paper should report agreement statistics (e.g., correlation or error between propagated and oracle scores) and describe how θ is chosen.
[Section 2.2, Phase 1] The coarse-ranking weights are computed using the Shapley value algorithm "based on analysis of historical dataset performance," but the historical data, the exact Shapley procedure, and the resulting weights are not described. This leaves a key free parameter that directly affects the ranking and all downstream selections. At minimum, the paper should disclose how the historical performance is defined, what features are used in the Shapley computation, and what the resulting weights are.

minor comments (4)

[Figure 2] Several mathematical symbols and letters render as invalid characters (e.g., "푖", "푆푖", "�") in the figure, making the formal definitions of scores and constraints unreadable. The figure should be regenerated with a proper math font or the notation should be defined in the caption or surrounding text.
[Section 1, Introduction] The phrase "such as OpenClaw and OpenAI Codex" includes a name that may be misspelled or refer to a nonstandard tool; please verify the intended agent names and cite the corresponding systems.
[Section 2.2, Phase 3] The knapsack formulation is described as a "two-dimensional knapsack problem" with budget and size constraints. It would be helpful to explicitly state that the dynamic programming runs over discretized price units and to give the time complexity, as this is a core algorithmic component.
[References] Some references lack access dates for web sources, and the artifact URL should be verified to be publicly accessible before the camera-ready deadline.

Circularity Check

0 steps flagged · score 0.0 of 10

No significant circularity: the paper is a system/demo description with no derivation that reduces to its own inputs, though its central empirical claim is unverified.

full rationale

Guixu is presented as an end-to-end demonstration system rather than a paper with a formal derivation. The valuation pipeline (multi-signal coarse ranking, proxy-label propagation, multi-round knapsack optimization) is a design proposal, not a theorem. The Shapley weights are said to be computed 'based on analysis of historical dataset performance,' but the paper does not specify what that analysis is, what data it uses, or how the resulting weights are evaluated; without that specification there is no exhibited reduction of a prediction to the fitted input. The strongest claim, that 'valuation-driven selection yields a measurable accuracy gain even at zero budget' (Section 3, Scenario 1), is an assertion about a live demonstration, not a derived or measured result. An unmeasured claim is a verifiability gap, not circularity. Likewise, the proxy-label propagation design 'yields substantial cost savings without sacrificing scoring fidelity' is an unvalidated assumption; no experiment compares propagated scores to oracle scores. That is missing evidence, not a self-referential derivation. The on-chain attestation signals feed back into valuation, forming a feedback loop by design, but the paper does not claim to derive trustworthiness from the same valuation scores it uses; it merely asserts that such signals 'can substantially improve data valuation.' There are no load-bearing self-citations, no uniqueness theorems imported from the authors, and no renamed known result presented as a first-principles derivation. Therefore no circular step can be quoted and exhibited, and the appropriate circularity score is 0.

Assumptions & free parameters 3 free parameters · 5 assumptions · 0 invented entities

The central system rests on several untested domain assumptions, particularly that proxy-label propagation and on-chain reputation inputs improve valuation quality. The only fitted parameters mentioned are the Shapley-derived signal weights, whose values are unspecified. No new theoretical entities are introduced.

free parameters (3)

coarse-ranking signal weights = not specified
Phase 1 combines six signals (relevance, schema fit, data scale, label quality, metadata completeness, on-chain reputation) with weights computed via Shapley from historical dataset performance. No values or validation are reported.
anchor score thresholds = not specified
Phase 2 partitions scored seeds into high-score and low-score anchors using configurable thresholds. Defaults are omitted.
propagation threshold theta = not specified
Phase 2 uses a configurable threshold to decide whether a sample inherits the mean anchor score or is re-judged by the LLM oracle. The value controls the cost-fidelity tradeoff but is not given.

assumptions (5)

domain assumption On-chain reputation signals (transaction volume, reviews, seller reputation) improve task-aware data valuation.
Section 2.3 feeds attestation signals into the valuation module, but the paper provides no evidence that these signals are predictive of data utility.
domain assumption Proxy-label propagation via textual similarity to anchor sets preserves LLM oracle scoring fidelity.
Section 2.2, Phase 2 assumes that samples similar to high-score anchors in embedding space have similar utility. This is plausible but untested.
domain assumption Zero-temperature, JSON-constrained LLM calls reliably extract task and data schemas from natural-language queries.
Section 2.2 NL Query Parser relies on LLM extraction without evaluating extraction accuracy.
standard math The 0-1 knapsack DP gives an optimal portfolio for the discretized budget and size constraints.
Standard dynamic programming for two-dimensional knapsack; not a new mathematical claim.
domain assumption Shamir 3-of-5 secret sharing via Lit Protocol ensures that no single party holds the complete decryption key.
Section 2.3 relies on the security properties of Shamir secret sharing, a standard cryptographic technique.

how reviews work

0 comments

Cite this review

Pith. "Pith review of Guixu: Valuation-Driven Data Discovery for Autonomous AI Agents with On-Chain Attestation." pith.science (2026). https://pith.science/paper/D5M4T7LT

@misc{pith2026260807949,
  author       = {Pith},
  title        = {Pith review of: Guixu: Valuation-Driven Data Discovery for Autonomous AI Agents with On-Chain Attestation},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/D5M4T7LT}},
  note         = {Machine review of arXiv:2608.07949}
}

read the original abstract

Autonomous agents increasingly rely on external data to complete downstream tasks such as model training and decision support. However, existing data discovery systems remain largely retrieval-oriented: they surface candidate datasets from heterogeneous sources, but provide limited support for estimating task-specific utility, selecting cost-effective datasets under budget constraints, or incorporating trustworthy feedback from prior usage. This paper presents Guixu, a valuation-driven data discovery system for autonomous agents. Guixu employs a three-phase valuation pipeline with proxy-label propagation and multi-round knapsack optimization for task-aware data valuation. Guixu integrates agentic payment protocol to enable budget-constrained data procurement workflows. Guixu leverages on-chain data market and attestation signals for verifiable data discovery. Our demonstration highlights how Guixu enables an agent to move beyond keyword-based dataset retrieval toward task- and budget-aware, trustworthy data discovery and procurement. Attendees can interactively explore the full workflow, from NL task specification and multi-source search to data valuation and verifiable transaction feedback.

Figures

Figures reproduced from arXiv: 2608.07949 by the authors.

**Figure 1.** Overview of the Guixu workflow. A user’s naturallanguage query is processed through six phases. Its importance is amplified in agentic settings, where autonomous agents such as OpenClaw and OpenAI Codex must independently locate, evaluate, and procure external data to complete complex tasks, including model training, prediction, and decision support. In such workflows, data discovery is not merely a retrieval probl… view at source ↗

**Figure 3.** End-to-end transaction lifecycle of the on-chain [PITH_FULL_IMAGE:figures/full_fig_p003_3.png] view at source ↗

**Figure 2.** Three-phase pipeline of task-aware data valuation. [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗

Figures from the paper (1 more)

**Figure 4.** Figure 4: The Graphical User Interface of Guixu. 3 DEMONSTRATION Demonstration Setup and Interaction. Guixu showcases an agent-native interface for valuation-driven data discovery and data acquisition for autonomous AI agents. The interface is connected to the Guixu MCP server, …

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

70 extracted references · 60 canonical work pages

[1]

Companion of the 2024 International Conference on Management of Data , pages =

Nathan, Vikram and Singh, Vikramank and Liu, Zhengchun and Rahman, Mohammad and Kipf, Andreas and Horn, Dominik and Pagano, Davide and Saxena, Gaurav and Narayanaswamy, Balakrishnan and Kraska, Tim , title =. Companion of the 2024 International Conference on Management of Data , pages =. 2024 , isbn =

work page 2024
[2]

Marcus, Ryan and Negi, Parimarjan and Mao, Hongzi and Zhang, Chi and Alizadeh, Mohammad and Kraska, Tim and Papaemmanouil, Olga and Tatbul, Nesime , title =. Proc. VLDB Endow. , month = jul, pages =. 2019 , issue_date =

work page 2019
[3]

SIGMOD Rec

Marcus, Ryan and Negi, Parimarjan and Mao, Hongzi and Tatbul, Nesime and Alizadeh, Mohammad and Kraska, Tim , title =. SIGMOD Rec. , month = jun, pages =. 2022 , issue_date =

work page 2022
[4]

Zhao, Yue and Cong, Gao and Shi, Jiachen and Miao, Chunyan , title =. Proc. VLDB Endow. , month = apr, pages =. 2022 , issue_date =

work page 2022
[5]

Zhu, Rong and Chen, Wei and Ding, Bolin and Chen, Xingguang and Pfadler, Andreas and Wu, Ziniu and Zhou, Jingren , title =. Proc. VLDB Endow. , month = feb, pages =. 2023 , issue_date =

work page 2023
[6]

Chen, Xu and Chen, Haitian and Liang, Zibo and Liu, Shuncheng and Wang, Jinghong and Zeng, Kai and Su, Han and Zheng, Kai , title =. Proc. VLDB Endow. , month = may, pages =. 2023 , issue_date =

work page 2023
[7]

Zhao, Yue and Li, Zhaodonghui and Cong, Gao , title =. Proc. VLDB Endow. , month = dec, pages =. 2023 , issue_date =

work page 2023
[8]

Proceedings of the VLDB Endowment , volume =

Viktor Leis and Andrey Gubichev and Atanas Mirchev and Peter Boncz and Alfons Kemper and Thomas Neumann , title =. Proceedings of the VLDB Endowment , volume =

work page

Show all 70 references

[9]

2024 , howpublished =

Amazon Web Services , title =. 2024 , howpublished =

2024
[10]

2024 , howpublished =

Snowflake , title =. 2024 , howpublished =

2024
[11]

2026 , howpublished =

Alibaba Cloud , title =. 2026 , howpublished =

2026
[12]

Workshop on Economics of Peer-to-Peer systems , volume=

Incentives build robustness in BitTorrent , author=. Workshop on Economics of Peer-to-Peer systems , volume=
[13]

2026 , howpublished =

OpenClaw , title =. 2026 , howpublished =

2026
[14]

2026 , howpublished =

Coinbase , title =. 2026 , howpublished =

2026
[15]

2026 , howpublished =

Stripe , title =. 2026 , howpublished =

2026
[16]

2026 , howpublished =

Lit Protocol , title =. 2026 , howpublished =

2026
[17]

CoRR , volume =

Juan Benet , title =. CoRR , volume =. 2014 , eprinttype =. 1407.3561 , timestamp =

2014 arXiv
[18]

2026 , howpublished =

OpenAI , title =. 2026 , howpublished =

2026
[19]

2026 , howpublished =

Kaggle , title =. 2026 , howpublished =

2026
[20]

2026 , howpublished =

Amazon Web Services , title =. 2026 , howpublished =

2026
[21]

Leonid Hurwicz , title =
[22]

Proceedings of the VLDB Endowment , volume =

Benjamin Hilprecht and Andreas Schmidt and Moritz Kulessa and Alejandro Molina and Kristian Kersting and Carsten Binnig , title =. Proceedings of the VLDB Endowment , volume =
[23]

arXiv preprint arXiv:2110.07875 , year =

Vijay Prakash Dwivedi and Anh Tuan Luu and Thomas Laurent and Yoshua Bengio and Xavier Bresson , title =. arXiv preprint arXiv:2110.07875 , year =

arXiv
[24]

Misegiannis, Michail Georgoulakis and Ritter, Daniel and Leis, Viktor and Giceva, Jana , title =. Proc. VLDB Endow. , month = jan, pages =. 2026 , issue_date =

2026
[25]

Neural Computation , volume =

Mikhail Belkin and Partha Niyogi , title =. Neural Computation , volume =
[26]

2023 , eprint=

Cost-Intelligent Data Analytics in the Cloud , author=. 2023 , eprint=

2023
[27]

2026 , eprint=

SafeLoad: Efficient Admission Control Framework for Identifying Memory-Overloading Queries in Cloud Data Warehouses , author=. 2026 , eprint=

2026
[28]

Lyu, Chenghao and Fan, Qi and Guyard, Philippe and Diao, Yanlei , title =. Proc. VLDB Endow. , month = jul, pages =. 2024 , issue_date =

2024
[29]

Rieger, Maximilian and Neumann, Thomas , title =. Proc. ACM Manag. Data , month = jun, articleno =. 2025 , issue_date =

2025
[30]

Lyu, Chenghao and Lachaud, Guillaume and Lozano, Gabriel and Diao, Yanlei , title =. Proc. VLDB Endow. , month = jan, pages =. 2026 , issue_date =

2026
[31]

Q-Cop: Avoiding bad query mixes to minimize client timeouts under heavy loads , booktitle =

Sean Tozer and Tim Brecht and Ashraf Aboulnaga , editor =. Q-Cop: Avoiding bad query mixes to minimize client timeouts under heavy loads , booktitle =
[32]

ActiveSLA: a profit-oriented admission control framework for database-as-a-service providers , booktitle =

PengCheng Xiong and Yun Chi and Shenghuo Zhu and Jun'ichi Tatemura and Calton Pu and Hakan Hacig. ActiveSLA: a profit-oriented admission control framework for database-as-a-service providers , booktitle =
[33]

, author=

Improvements in beam search. , author=. ICSLP , volume=
[34]

2025 , howpublished =

The SafeLoad authors , title =. 2025 , howpublished =

2025
[35]

2026 , howpublished =

AWS , title =. 2026 , howpublished =

2026
[36]

2026 , howpublished =

MySQL , title =. 2026 , howpublished =

2026
[37]

2026 , howpublished =

BigQuery , title =. 2026 , howpublished =

2026
[38]

2026 , howpublished =

Snowflake , title =. 2026 , howpublished =

2026
[39]

The Annals of Mathematical Statistics , volume=

On estimation of a probability density function and mode , author=. The Annals of Mathematical Statistics , volume=. 1962 , publisher=

1962
[40]

Visualizing Data using t-SNE , volume =

van der Maaten, Laurens and Hinton, Geoffrey , year =. Visualizing Data using t-SNE , volume =
[41]

Companion of the 2024 International Conference on Management of Data , pages =

Wu, Ziniu and Marcus, Ryan and Liu, Zhengchun and Negi, Parimarjan and Nathan, Vikram and Pfeil, Pascal and Saxena, Gaurav and Rahman, Mohammad and Narayanaswamy, Balakrishnan and Kraska, Tim , title =. Companion of the 2024 International Conference on Management of Data , pag...

2024
[42]

Armenatzoglou, Nikos and Basu, Sanuj and Bhanoori, Naga and Cai, Mengchu and Chainani, Naresh and Chinta, Kiran and Govindaraju, Venkatraman and Green, Todd J. and Gupta, Monish and Hillig, Sebastian and Hotinger, Eric and Leshinksy, Yan and Liang, Jintian and McCreedy, Michae...

2022
[43]

Auto-WLM: Machine Learning Enhanced Workload Management in Amazon Redshift , booktitle =

Gaurav Saxena and Mohammad Rahman and Naresh Chainani and Chunbin Lin and George Caragea and Fahim Chowdhury and Ryan Marcus and Tim Kraska and Ippokratis Pandis and Balakrishnan (Murali) Narayanaswamy , editor =. Auto-WLM: Machine Learning Enhanced Workload Management in Amaz...
[44]

2025 , eprint=

How Good are Learned Cost Models, Really? Insights from Query Optimization Tasks , author=. 2025 , eprint=

2025
[45]

Advances in neural information processing systems , volume=

A unified approach to interpreting model predictions , author=. Advances in neural information processing systems , volume=
[46]

Colmenares , editor =

Hao Xu and Juan A. Colmenares , editor =. Bouncer: Admission Control with Response Time Objectives for Low-latency Online Data Systems , booktitle =
[47]

40th International Conference on Data Engineering,

Tim Zeyl and Harshwin Venugopal and Calvin Sun and Paul Larson , title =. 40th International Conference on Data Engineering,
[48]

The World Wide Web Conference , pages =

Brickley, Dan and Burgess, Matthew and Noy, Natasha , title =. The World Wide Web Conference , pages =. 2019 , isbn =

2019
[49]

and Chen, Jiaoyan and Wu, Zhenyu , title =

Paton, Norman W. and Chen, Jiaoyan and Wu, Zhenyu , title =. ACM Comput. Surv. , month = nov, articleno =. 2023 , issue_date =

2023
[50]

Chunxu Tang and Beinan Wang and Zhenxiao Luo and Huijun Wu and Shajan Dasan and Maosong Fu and Yao Li and Mainak Ghosh and Ruchin Kabra and Nikhil Kantibhai Navadiya and Da Cheng and Fred Dai and Vrushali Channapattan and Prachi Mishra , title =
[51]

CoRR , volume =

Shaikh Quader and Andres Jaramillo and Sumona Mukhopadhyay and Ghadeer Abuoda and Calisto Zuzarte and David Kalmuk and Marin Litoiu and Manos Papagelis , title =. CoRR , volume =. 2024 , eprinttype =

2024
[52]

Kuno and Umeshwar Dayal and Janet L

Archana Ganapathi and Harumi A. Kuno and Umeshwar Dayal and Janet L. Wiener and Armando Fox and Michael I. Jordan and David A. Patterson , title =. Proceedings of the 25th International Conference on Data Engineering,
[53]

Donghui Wang and Peng Cai and Weining Qian and Aoying Zhou , title =. 37th
[54]

arXiv preprint arXiv:2401.08281 , year=

The faiss library , author=. arXiv preprint arXiv:2401.08281 , year=

arXiv
[55]

Zhan, Chaoqun and Su, Maomeng and Wei, Chuangxian and Peng, Xiaoqiang and Lin, Liang and Wang, Sheng and Chen, Zhe and Li, Feifei and Pan, Yue and Zheng, Fang and Chai, Chengliang , title =. Proc. VLDB Endow. , month = aug, pages =. 2019 , issue_date =

2019
[56]

2026 , note =

AnalyticDB Compute Unit (ACU) Terms and Definitions , howpublished =. 2026 , note =

2026
[57]

2026 , note =

Amazon Redshift Serverless Capacity and RPUs , howpublished =. 2026 , note =

2026
[58]

Proceedings of the 14th USENIX Conference on Networked Systems Design and Implementation , pages =

Alipourfard, Omid and Liu, Hongqiang Harry and Chen, Jianshu and Venkataraman, Shivaram and Yu, Minlan and Zhang, Ming , title =. Proceedings of the 14th USENIX Conference on Networked Systems Design and Implementation , pages =. 2017 , isbn =

2017
[59]

and Motivala, Ashish and Munir, Abdul Q

Dageville, Benoit and Cruanes, Thierry and Zukowski, Marcin and Antonov, Vadim and Avanes, Artin and Bock, Jon and Claybaugh, Jonathan and Engovatov, Daniel and Hentschel, Martin and Huang, Jiansheng and Lee, Allison W. and Motivala, Ashish and Munir, Abdul Q. and Pelley, Stev...

2016
[60]

and Buyya, Rajkumar , title =

Qu, Chenhao and Calheiros, Rodrigo N. and Buyya, Rajkumar , title =. ACM Comput. Surv. , month = jul, articleno =. 2018 , issue_date =. doi:10.1145/3148149 , abstract =

2018 doi
[61]

Proceedings of the 2018 USENIX Conference on Usenix Annual Technical Conference , pages =

Li, Zhao Lucis and Liang, Chieh-Jan Mike and He, Wenjia and Zhu, Lianjie and Dai, Wenjun and Jiang, Jin and Sun, Guangzhong , title =. Proceedings of the 2018 USENIX Conference on Usenix Annual Technical Conference , pages =. 2018 , isbn =

2018
[62]

Melnik, Sergey and Gubarev, Andrey and Long, Jing Jing and Romer, Geoffrey and Shivakumar, Shiva and Tolton, Matt and Vassilakis, Theo and Ahmadi, Hossein and Delorey, Dan and Min, Slava and Pasumansky, Mosha and Shute, Jeff , title =. Proc. VLDB Endow. , month = aug, pages =....

2020
[63]

CoRR , volume =

Chujun Song and Slim Bouguerra and Erik Krogen and Daniel Abadi , title =. CoRR , volume =
[64]

Zibo Liang and Xu Chen and Yuyang Xia and Runfan Ye and Haitian Chen and Jiandong Xie and Kai Zheng , title =. 40th
[65]

Proceedings of the 2020 ACM SIGMOD International Conference on Management of Data , pages =

Rzadca, Krzysztof and Findeisen, Pawel and Swiderski, Jacek and Zych, Przemyslaw and Broniek, Przemyslaw and Kusmierek, Jacek and Gorczyk, Pawel and Liu, Xiao and Wiggers, Gregory and Magid, Itai and others , title =. Proceedings of the 2020 ACM SIGMOD International Conference...

2020
[66]

Proceedings of the 26th Symposium on Operating Systems Principles (SOSP) , pages =

Cortez, Eli and Bonde, Anand and Muzio, Benjamin and Russinovich, Mark and Fontoura, Marcus and Bianchini, Ricardo , title =. Proceedings of the 26th Symposium on Operating Systems Principles (SOSP) , pages =
[67]

2019 , url =

Trent McConaghy and others , title =. 2019 , url =

2019
[68]

2026 , author =

Guixu Hub: The Data Discovery and Market Platform for Autonomous AI Agents , url =. 2026 , author =

2026
[69]

Companion of the 2025 International Conference on Management of Data (SIGMOD/PODS) , year =

Daniel Alabi and Sainyam Galhotra and Shagufta Mehnaz and Zeyu Song and Eugene Wu , title =. Companion of the 2025 International Conference on Management of Data (SIGMOD/PODS) , year =

2025
[70]

Zhang, Zhuo and Li, Chao and Tao, Yangyu and Yang, Renyu and Tang, Hong and Xu, Jie , title =. Proc. VLDB Endow. , month = aug, pages =. 2014 , issue_date =. doi:10.14778/2733004.2733012 , abstract =

2014

Pith tools

Reviewed August 12, 2026 · model on record in the stance chip above.

[1] [1]

Companion of the 2024 International Conference on Management of Data , pages =

Nathan, Vikram and Singh, Vikramank and Liu, Zhengchun and Rahman, Mohammad and Kipf, Andreas and Horn, Dominik and Pagano, Davide and Saxena, Gaurav and Narayanaswamy, Balakrishnan and Kraska, Tim , title =. Companion of the 2024 International Conference on Management of Data , pages =. 2024 , isbn =

work page 2024

[2] [2]

Marcus, Ryan and Negi, Parimarjan and Mao, Hongzi and Zhang, Chi and Alizadeh, Mohammad and Kraska, Tim and Papaemmanouil, Olga and Tatbul, Nesime , title =. Proc. VLDB Endow. , month = jul, pages =. 2019 , issue_date =

work page 2019

[3] [3]

SIGMOD Rec

Marcus, Ryan and Negi, Parimarjan and Mao, Hongzi and Tatbul, Nesime and Alizadeh, Mohammad and Kraska, Tim , title =. SIGMOD Rec. , month = jun, pages =. 2022 , issue_date =

work page 2022

[4] [4]

Zhao, Yue and Cong, Gao and Shi, Jiachen and Miao, Chunyan , title =. Proc. VLDB Endow. , month = apr, pages =. 2022 , issue_date =

work page 2022

[5] [5]

Zhu, Rong and Chen, Wei and Ding, Bolin and Chen, Xingguang and Pfadler, Andreas and Wu, Ziniu and Zhou, Jingren , title =. Proc. VLDB Endow. , month = feb, pages =. 2023 , issue_date =

work page 2023

[6] [6]

Chen, Xu and Chen, Haitian and Liang, Zibo and Liu, Shuncheng and Wang, Jinghong and Zeng, Kai and Su, Han and Zheng, Kai , title =. Proc. VLDB Endow. , month = may, pages =. 2023 , issue_date =

work page 2023

[7] [7]

Zhao, Yue and Li, Zhaodonghui and Cong, Gao , title =. Proc. VLDB Endow. , month = dec, pages =. 2023 , issue_date =

work page 2023

[8] [8]

Proceedings of the VLDB Endowment , volume =

Viktor Leis and Andrey Gubichev and Atanas Mirchev and Peter Boncz and Alfons Kemper and Thomas Neumann , title =. Proceedings of the VLDB Endowment , volume =

work page