arXiv subjects

Raul Castro Fernandez

Publications and source records attributed to Raul Castro Fernandez.

At least 19 recordsLinked to original sources

arXiv2026

Enabling Personal Dataflow Sovereignty via Bolt-on Data Escrow

The digital economy is powered by a continuous and massive exchange of personal data. Individuals provide data to platforms in return for services, from social networking and search to health monitoring, entertainment, and access to LLMs. This exchange has created immense value, but it has also established a fundamental asymmetry of power: individuals possess only coarse-grained control over data access rather than fine-grained control over its purpose of use, creating a gap where data can be repurposed for undisclosed uses, e.g., platforms selling the data to data brokers, which results in a critical loss of personal data sovereignty. This paper frames this socio-technical challenge as a dataflow management problem. We propose a bolt-on data escrow architecture through delegated computation. In our model, instead of data flowing to platforms, platforms delegate their computation to a trustworthy escrow. This inversion empowers individuals with transparency and control over their dataflows. We present four contributions: (1) a dataflow model that explicitly incorporates computational purpose as a first-class primitive; (2) a minimally invasive programming interface, run(access(), compute()), built on a unified relational interface that virtualizes on-device data sources and a computation offloading component; (3) a concrete implementation of our escrow within the Apple ecosystem, demonstrating its practicality; and (4) both qualitative and quantitative evaluations demonstrating that our solution is expressive enough to implement a wide range of dataflows from real-world applications and introduces minimal runtime overhead. In summary, our work serves as a stepping stone toward achieving personal dataflow sovereignty.

cs.DB↗

arXiv2026

Within-Dataset Disclosure Risk for Differential Privacy

Differential privacy (DP) enables private data analysis. In a typical DP deployment, controllers manage individuals' sensitive data and are responsible for answering analysts' queries while protecting individuals' privacy. They do so by choosing the privacy parameter $ε$, which controls the degree of privacy for all individuals in all possible datasets. However, it is challenging for controllers to choose $ε$ because of the difficulty of interpreting the privacy implications of such a choice on the within-dataset individuals. To address this challenge, we first derive a relative disclosure risk indicator (RDR) that indicates the impact of choosing $ε$ on the within-dataset individuals' disclosure risk. We then design an algorithm to find $ε$ based on controllers' privacy preferences expressed as a function of the within-dataset individuals' RDRs, and an alternative algorithm that finds and releases $ε$ while satisfying DP. Lastly, we propose a solution that bounds the total privacy leakage when using the algorithm to answer multiple queries without requiring controllers to set the total privacy budget. We evaluate our contributions through an IRB-approved user study that shows the RDR is useful for helping controllers choose $ε$, and experimental evaluations showing our algorithms are efficient and scalable.

cs.DB↗

arXiv2026

Demonstration of Pneuma-Seeker: Agentic System for Reifying and Fulfilling Information Needs on Tabular Data

Data analysts working with relational data often start with vague or underspecified questions and refine them iteratively as they explore the data. To support this iterative process, we demonstrate Pneuma-Seeker, a system that reifies a user's information need as explicit, inspectable relational specifications, enabling iterative refinement of the information need, targeted data discovery, and provenance-aware execution. Through two real-world procurement use cases, we show how Pneuma-Seeker leverages LLMs as transparent, interactive analytical collaborators rather than opaque answer engines.

cs.AI↗

arXiv2026

Pneuma-Seeker: A Relational Reification Mechanism to Align AI Agents with Human Work over Relational Data

When faced with data problems, many data workers cannot articulate their information need precisely enough for software to help. Although LLMs interpret natural-language requests, they behave brittly when intent is under-specified, e.g., hallucinating fields, assuming join paths, or producing ungrounded answers. We present Pneuma-Seeker, a system built around a central idea: relational reification. Pneuma-Seeker represents a user's evolving information need as a relational schema: a concrete, analysis-ready data model shared between user and system. Rather than answering prompts directly, Pneuma-Seeker iteratively refines this schema, then discovers and prepares relevant sources to construct a relation and executable program that compute the answer. Pneuma-Seeker employs an LLM-powered agentic architecture with conductor-style planning and macro- and micro-level context management to operate effectively over heterogeneous relational corpora. We evaluate Pneuma-Seeker across multiple domains against state-of-the-art academic and industrial baselines, demonstrating higher answer accuracy. Deployment in a real organization highlights trust and inspectability as essential requirements for LLM-mediated data systems.

cs.DB↗

arXiv2026

The Pneuma Project: Reifying Information Needs as Relational Schemas to Automate Discovery, Guide Preparation, and Align Data with Intent

Data discovery and preparation remain persistent bottlenecks in the data management lifecycle, especially when user intent is vague, evolving, or difficult to operationalize. The Pneuma Project introduces Pneuma-Seeker, a system that helps users articulate and fulfill information needs through iterative interaction with a language model-powered platform. The system reifies the user's evolving information need as a relational data model and incrementally converges toward a usable document aligned with that intent. To achieve this, the system combines three architectural ideas: context specialization to reduce LLM burden across subtasks, a conductor-style planner to assemble dynamic execution plans, and a convergence mechanism based on shared state. The system integrates recent advances in retrieval-augmented generation (RAG), agentic frameworks, and structured data preparation to support semi-automatic, language-guided workflows. We evaluate the system through LLM-based user simulations and show that it helps surface latent intent, guide discovery, and produce fit-for-purpose documents. It also acts as an emergent documentation layer, capturing institutional knowledge and supporting organizational memory.

cs.DB↗

arXiv2025

Core Hours and Carbon Credits: Incentivizing Sustainability in HPC

Realizing a shared responsibility between providers and consumers is critical to manage the sustainability of HPC. However, while cost may motivate efficiency improvements by infrastructure operators, broader progress is impeded by a lack of user incentives. We conduct a survey of HPC users that reveals fewer than 30 percent are aware of their energy consumption, and that energy efficiency is among users' lowest priority concerns. One explanation is that existing pricing models may encourage users to prioritize performance over energy efficiency. We propose two transparent multi-resource pricing schemes, Energy- and Carbon-Based Accounting, that seek to change this paradigm by incentivizing more efficient user behavior. These two schemes charge for computations based on their energy consumption or carbon footprint, respectively, rewarding users who leverage efficient hardware and software. We evaluate these two pricing schemes via simulation, in a prototype, and a user study.

cs.DC↗

arXiv2025

Emergent evaluation hubs in a decentralizing large language model ecosystem

Large language models are proliferating, and so are the benchmarks that serve as their common yardsticks. We ask how the agglomeration patterns of these two layers compare: do they evolve in tandem or diverge? Drawing on two curated proxies for the ecosystem, the Stanford Foundation-Model Ecosystem Graph and the Evidently AI benchmark registry, we find complementary but contrasting dynamics. Model creation has broadened across countries and organizations and diversified in modality, licensing, and access. Benchmark influence, by contrast, displays centralizing patterns: in the inferred benchmark-author-institution network, the top 15% of nodes account for over 80% of high-betweenness paths, three countries produce 83% of benchmark outputs, and the global Gini for inferred benchmark authority reaches 0.89. An agent-based simulation highlights three mechanisms: higher entry of new benchmarks reduces concentration; rapid inflows can temporarily complicate coordination in evaluation; and stronger penalties against over-fitting have limited effect. Taken together, these results suggest that concentrated benchmark influence functions as coordination infrastructure that supports standardization, comparability, and reproducibility amid rising heterogeneity in model production, while also introducing trade-offs such as path dependence, selective visibility, and diminishing discriminative power as leaderboards saturate.

cs.CY↗

arXiv2025

Where Does Academic Database Research Go From Here?

Panel proposal for an open forum to discuss and debate the future of database research in the context of industry, other research communities, and AI. Includes summaries of past panels, positions from panelists, as well as positions from a sample of the data management community.

cs.DB↗

arXiv2025

Mass-Scale Analysis of In-the-Wild Conversations Reveals Complexity Bounds on LLM Jailbreaking

As large language models (LLMs) become increasingly deployed, understanding the complexity and evolution of jailbreaking strategies is critical for AI safety. We present a mass-scale empirical analysis of jailbreak complexity across over 2 million real-world conversations from diverse platforms, including dedicated jailbreaking communities and general-purpose chatbots. Using a range of complexity metrics spanning probabilistic measures, lexical diversity, compression ratios, and cognitive load indicators, we find that jailbreak attempts do not exhibit significantly higher complexity than normal conversations. This pattern holds consistently across specialized jailbreaking communities and general user populations, suggesting practical bounds on attack sophistication. Temporal analysis reveals that while user attack toxicity and complexity remains stable over time, assistant response toxicity has decreased, indicating improving safety mechanisms. The absence of power-law scaling in complexity distributions further points to natural limits on jailbreak development. Our findings challenge the prevailing narrative of an escalating arms race between attackers and defenders, instead suggesting that LLM safety evolution is bounded by human ingenuity constraints while defensive measures continue advancing. Our results highlight critical information hazards in academic jailbreak disclosure, as sophisticated attacks exceeding current complexity baselines could disrupt the observed equilibrium and enable widespread harm before defensive adaptation.

cs.CL↗

arXiv2025

Exploiting LLMs for Automatic Hypothesis Assessment via a Logit-Based Calibrated Prior

As hypothesis generation becomes increasingly automated, a new bottleneck has emerged: hypothesis assessment. Modern systems can surface thousands of statistical relationships-correlations, trends, causal links-but offer little guidance on which ones are novel, non-trivial, or worthy of expert attention. In this work, we study the complementary problem to hypothesis generation: automatic hypothesis assessment. Specifically, we ask: given a large set of statistical relationships, can we automatically assess which ones are novel and worth further exploration? We focus on correlations as they are a common entry point in exploratory data analysis that often serve as the basis for forming deeper scientific or causal hypotheses. To support automatic assessment, we propose to leverage the vast knowledge encoded in LLMs' weights to derive a prior distribution over the correlation value of a variable pair. If an LLM's prior expects the correlation value observed, then such correlation is not surprising, and vice versa. We propose the Logit-based Calibrated Prior, an LLM-elicited correlation prior that transforms the model's raw output logits into a calibrated, continuous predictive distribution over correlation values. We evaluate the prior on a benchmark of 2,096 real-world variable pairs and it achieves a sign accuracy of 78.8%, a mean absolute error of 0.26, and 95% credible interval coverage of 89.2% in predicting Pearson correlation coefficient. It also outperforms a fine-tuned RoBERTa classifier in binary correlation prediction and achieves higher precision@K in hypothesis ranking. We further show that the prior generalizes to correlations not seen during LLM pretraining, reflecting context-sensitive reasoning rather than memorization.

cs.LG↗

arXiv2025

Optimal Pricing for Data-Augmented AutoML Marketplaces

Organizations often lack sufficient data to effectively train machine learning (ML) models, while others possess valuable data that remains underutilized. Data markets promise to unlock substantial value by matching data suppliers with demand from ML consumers. However, market design involves addressing intricate challenges, including data pricing, fairness, robustness, and strategic behavior. In this paper, we propose a pragmatic data-augmented AutoML market that seamlessly integrates with existing cloud-based AutoML platforms such as Google's Vertex AI and Amazon's SageMaker. Unlike standard AutoML solutions, our design automatically augments buyer-submitted training data with valuable external datasets, pricing the resulting models based on their measurable performance improvements rather than computational costs as the status quo. Our key innovation is a pricing mechanism grounded in the instrumental value - the marginal model quality improvement - of externally sourced data. This approach bypasses direct dataset pricing complexities, mitigates strategic buyer behavior, and accommodates diverse buyer valuations through menu-based options. By integrating automated data and model discovery, our solution not only enhances ML outcomes but also establishes an economically sustainable framework for monetizing external data.

cs.LG↗

arXiv2025

Pneuma: Leveraging LLMs for Tabular Data Representation and Retrieval in an End-to-End System

Finding relevant tables among databases, lakes, and repositories is the first step in extracting value from data. Such a task remains difficult because assessing whether a table is relevant to a problem does not always depend only on its content but also on the context, which is usually tribal knowledge known to the individual or team. While tools like data catalogs and academic data discovery systems target this problem, they rely on keyword search or more complex interfaces, limiting non-technical users' ability to find relevant data. The advent of large language models (LLMs) offers a unique opportunity for users to ask questions directly in natural language, making dataset discovery more intuitive, accessible, and efficient. In this paper, we introduce Pneuma, a retrieval-augmented generation (RAG) system designed to efficiently and effectively discover tabular data. Pneuma leverages large language models (LLMs) for both table representation and table retrieval. For table representation, Pneuma preserves schema and row-level information to ensure comprehensive data understanding. For table retrieval, Pneuma augments LLMs with traditional information retrieval techniques, such as full-text and vector search, harnessing the strengths of both to improve retrieval performance. To evaluate Pneuma, we generate comprehensive benchmarks that simulate table discovery workload on six real-world datasets including enterprise data, scientific databases, warehousing data, and open data. Our results demonstrate that Pneuma outperforms widely used table search systems (such as full-text search and state-of-the-art RAG systems) in accuracy and resource efficiency.

cs.DB↗

arXiv2024

A Survey on Data Markets

Data is the new oil of the 21st century. The growing trend of trading data for greater welfare has led to the emergence of data markets. A data market is any mechanism whereby the exchange of data products including datasets and data derivatives takes place as a result of data buyers and data sellers being in contact with one another, either directly or through mediating agents. It serves as a coordinating mechanism by which several functions, including the pricing and the distribution of data as the most important ones, interact to make the value of data fully exploited and enhanced. In this article, we present a comprehensive survey of this important and emerging direction from the aspects of data search, data productization, data transaction, data pricing, revenue allocation as well as privacy, security, and trust issues. We also investigate the government policies and industry status of data markets across different countries and different domains. Finally, we identify the unresolved challenges and discuss possible future directions for the development of data markets.

cs.GT↗

arXiv2024

FabricQA-Extractor: A Question Answering System to Extract Information from Documents using Natural Language Questions

Reading comprehension models answer questions posed in natural language when provided with a short passage of text. They present an opportunity to address a long-standing challenge in data management: the extraction of structured data from unstructured text. Consequently, several approaches are using these models to perform information extraction. However, these modern approaches leave an opportunity behind because they do not exploit the relational structure of the target extraction table. In this paper, we introduce a new model, Relation Coherence, that exploits knowledge of the relational structure to improve the extraction quality. We incorporate the Relation Coherence model as part of FabricQA-Extractor, an end-to-end system we built from scratch to conduct large scale extraction tasks over millions of documents. We demonstrate on two datasets with millions of passages that Relation Coherence boosts extraction performance and evaluate FabricQA-Extractor on large scale datasets.

cs.IR↗

arXiv2024

Programmable Dataflows: Abstraction and Programming Model for Data Sharing

Data sharing is central to a wide variety of applications such as fraud detection, ad matching, and research. The lack of data sharing abstractions makes the solution to each data sharing problem bespoke and cost-intensive, hampering value generation. In this paper, we first introduce a data sharing model to represent every data sharing problem with a sequence of dataflows. From the model, we distill an abstraction, the contract, which agents use to communicate the intent of a dataflow and evaluate its consequences, before the dataflow takes place. This helps agents move towards a common sharing goal without violating any regulatory and privacy constraints. Then, we design and implement the contract programming model (CPM), which allows agents to program data sharing applications catered to each problem's needs. Contracts permit data sharing, but their interactive nature may introduce inefficiencies. To mitigate those inefficiencies, we extend the CPM so that it can save intermediate outputs of dataflows, and skip computation if a dataflow tries to access data that it does not have access to. In our evaluation, we show that 1) the contract abstraction is general enough to represent a wide range of sharing problems, 2) we can write programs for complex data sharing problems and exhibit qualitative improvements over other alternate technologies, and 3) quantitatively, our optimizations make sharing programs written with the CPM efficient.

cs.DB↗

arXiv2024

Saving Money for Analytical Workloads in the Cloud

As users migrate their analytical workloads to cloud databases, it is becoming just as important to reduce monetary costs as it is to optimize query runtime. In the cloud, a query is billed based on either its compute time or the amount of data it processes. We observe that analytical queries are either compute- or IO-bound and each query type executes cheaper in a different pricing model. We exploit this opportunity and propose methods to build cheaper execution plans across pricing models that complete within user-defined runtime constraints. We implement these methods and produce execution plans spanning multiple pricing models that reduce the monetary cost for workloads by as much as 56%. We reduce individual query costs by as much as 90%. The prices chosen by cloud vendors for cloud services also impact savings opportunities. To study this effect, we simulate our proposed methods with different cloud prices and observe that multi-cloud savings are robust to changes in cloud vendor prices. These results indicate the massive opportunity to save money by executing workloads across multiple pricing models.

cs.DB↗

arXiv2023

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

Most deployed data discovery systems, such as Google Datasets, and open data portals only support keyword search. Keyword search is geared towards general audiences but limits the types of queries the systems can answer. We propose a new system that lets users write natural language questions directly. A major barrier to using this learned data discovery system is it needs expensive-to-collect training data, thus limiting its utility. In this paper, we introduce a self-supervised approach to assemble training datasets and train learned discovery systems without human intervention. It requires addressing several challenges, including the design of self-supervised strategies for data discovery, table representation strategies to feed to the models, and relevance models that work well with the synthetically generated questions. We combine all the above contributions into a system, Solo, that solves the problem end to end. The evaluation results demonstrate the new techniques outperform state-of-the-art approaches on well-known benchmarks. All in all, the technique is a stepping stone towards building learned discovery systems. The code is open-sourced at https://github.com/TheDataStation/solo

cs.IR↗

arXiv2023

Saibot: A Differentially Private Data Search Platform

Recent data search platforms use ML task-based utility measures rather than metadata-based keywords, to search large dataset corpora. Requesters submit a training dataset and these platforms search for augmentations (join or union compatible datasets) that, when used to augment the requester's dataset, most improve model (e.g., linear regression) performance. Although effective, providers that manage personally identifiable data demand differential privacy (DP) guarantees before granting these platforms data access. Unfortunately, making data search differentially private is nontrivial, as a single search can involve training and evaluating datasets hundreds or thousands of times, quickly depleting privacy budgets. We present Saibot, a differentially private data search platform that employs Factorized Privacy Mechanism (FPM), a novel DP mechanism, to calculate sufficient semi-ring statistics for ML over different combinations of datasets. These statistics are privatized once, and can be freely reused for the search. This allows Saibot to scale to arbitrary numbers of datasets and requests, while minimizing the amount that DP noise affects search results. We optimize the sensitivity of FPM for common augmentation operations, and analyze its properties with respect to linear regression. Specifically, we develop an unbiased estimator for many-to-many joins, prove its bounds, and develop an optimization to redistribute DP noise to minimize the impact on the model. Our evaluation on a real-world dataset corpus of 329 datasets demonstrates that Saibot can return augmentations that achieve model accuracy within 50 to 90% of non-private search, while the leading alternative DP mechanisms (TPM, APM, shuffling) are several orders of magnitude worse.

cs.DB↗