arXiv subjects

Eugene Wu

Publications and source records attributed to Eugene Wu.

At least 19 recordsLinked to original sources

arXiv2026

FlowCheck: Helping End-Users Specify and Verify Intent in Vibe-Coded Web Apps

Vibe-coded applications often contain silent behavioral failures in which the interface appears functional even though user-visible information does not flow to the expected state or output. We introduce FlowCheck, a constraint language to specify these user-visible information flows directly through the application interface, where constraints can also be displayed and inspected without reading code, and are structured enough for reliable LLM generation. FlowCheck translates the constraints into deterministic CodeQL analyses, and we evaluate it across four applications generated via Claude Code, and compare with three coding models as bug-finding baselines. We find that FlowCheck correctly translates and flags all 30 of our injected constraint violations with no false positives. In contrast, frontier models (Claude Opus 4.7, DeepSeek V3, and Gemini Pro) showed significantly lower accuracy when prompted to find bugs in the same code, with none achieving full accuracy. This approach lets vibe coders state intent in terms of the interface they understand, and checks it deterministically against the code they do not.

cs.SE↗

arXiv2026

Agentic Data Environments

Autonomous agents promise substantial gains in speed, scale, and labor efficiency, but their failures can impose abrupt and often irreversible costs. The central challenge for agentic automation is therefore to increase the benefits of automation while bounding the consequences of failure. While databases remain central to modern computing, agents operate over a broader data environment spanning files, APIs, applications, and system state. In this talk, I will outline early work on Agentic Data Environments -- the execution substrate in which agents operate -- that both amplify agent capabilities and enforce safety guarantees. This perspective reframes data systems from passive stores of state into active substrates for safe, reliable execution.

cs.AI↗

arXiv2026

SANA: What Matters for QA Agents over Massive Data Lakes?

Exploratory question answering (EQA) over data lakes requires an LLM agent to discover relevant sources, analyze retrieved data, and adapt its actions based on intermediate results. End-to-end accuracy alone cannot distinguish failures in search, planning, data analysis, or the agent's Action Policy: its decisions about what to do next and when to submit an answer. We present SANA (Search Agent Navigation Ablation framework), a diagnostic ablation framework that transforms EQA tasks into runtime profiles containing gold source sequence, sanitized subquestions, and execution records. SANA uses these profiles to construct idealized search, planning, and data-analysis tools, allowing each component to be ablated; the residual gap is diagnostic evidence for policy failures. To illustrate SANA as a reusable evaluation framework, we adapted two recent EQA benchmarks, LakeQA and KramaBench, and evaluated lightweight and mid-sized agents under fixed prompts, budgets, data lakes, and runtimes. Across both benchmarks, data analysis is a consistent bottleneck while planning is less so. Search is a major limitation in LakeQA's large data-lake setting, but less so for the smaller-scale KramaBench. SANA thus deconstructs end-to-end task accuracies into a diagnosis of where data-lake agents fail, and allows for systematic comparisons of progress in search, planning, data analysis, and agent design.

cs.CL↗

arXiv2026

LakeQA: An Exploratory QA Benchmark over a Million-Scale Data Lake

Recent large language models (LLMs) have shown rapid progress in reading-based question answering (QA), where evidence is explicitly provided or can be trivially retrieved. In contrast, real-world questions are often not paired with accurate evidence documents. The useful evidence resides in massive data lakes, making search a prerequisite for answering. However, there is a lack of comprehensive benchmarks that require both searching and reasoning over large data lakes. To this end, we introduce LakeQA, a comprehensive benchmark for search-centric question answering over data lakes that jointly emphasizes searching and reasoning capabilities. LakeQA is built on a heterogeneous collection of approximately 9.5 TB of text resources from Wikipedia and open-source government data, spanning structured and unstructured data. To ensure task quality, each sample is annotated by at least one Ph.D.-level expert. Each task requires long-horizon multi-hop reasoning with implicit intermediate steps: agents need to discover the correct documents and then compose evidence across sources to produce the answer. Experimental results on seven frontier LLMs demonstrate that LakeQA is challenging. For instance, GPT-5.2 achieves only an exact-match score of 18.37% on LakeQA. Overall, LakeQA provides a realistic testbed for developing LLM agents that can both find and analyze data in modern data lakes.

cs.CL↗

arXiv2026

Data Flow Control: Data Safety Policies for AI Agents

Agents increasingly generate SQL, orchestrate pipelines, and automate data analysis on behalf of users. While recent work improves query correctness, correctness is not safety. A query may be semantically valid yet violate regulatory, privacy, or business constraints that govern how data may be combined and released. We argue that enforcing such constraints is fundamentally a data infrastructure problem. This paper introduces Data Flow Control (DFC), a framework to declaratively specify and guarantee policy enforcement over tuple-level data flows within a DBMS query. A key challenge is defining a policy language that is optimizer-invariant yet efficient to enforce at scale. We formalize data safety as aggregate predicates over provenance monomials and present Passant, a portable query rewriting layer that enforces DFC policies without materializing provenance. Across five DBMS engines -- DuckDB, Umbra, PostgreSQL, DataFusion, and SQLServer -- Passant achieves ~0% overhead and outperforms alternatives by orders of magnitude. As a result, Data Flow Control is the first step towards moving data safety from prompts and post-hoc checks into the data infrastructure. Data Flow Control is available open source at https://github.com/dataflowcontrol/data-flow-control.

cs.DB↗

arXiv2026

Latent Cache Flow: Model-to-Model Communication Without Text

LLM agents today communicate via text, which incurs considerable latency and information loss due to the need to autoregressively decode the sharer model's state and encode at the receiver model. Recent work such as Cache-to-Cache (C2C; Fu et al., 2026) seeks to exchange KV caches by learning adapters that translate sharer KV matrices to the receiver model. However, the adapters are large and expensive to train, and translate individual tokens, which requires the target context to be identical. This is unsuitable for agent communication, where the LLMs have differing context. We introduce Latent Cache Flow (LCF). To address efficiency, we observe that keys and values can be jointly translated and compressed, reducing the adapter to about 4% of C2C's size. To address differing context, we design the adapter to transmit a summary of new information that the target model does not have. Our early experiments show that a pruned 13 MB LCF adapter can be more accurate than C2C at 956 MB in shared-context settings; for different contexts, LCF improves F1 by 7.5% and Exact Match by 23% while 8.5 times faster than text-based communication.

cs.LG↗

arXiv2026

BranchBench: Aligning Database Branching with Agentic Demands

Branchable databases are evolving from developer tools to infrastructure for agentic workloads characterized by speculative mutations and non-linear state exploration. Traditional RDBMS mechanisms such as nested transactions do not provide the persistent isolation and concurrent branch management required by autonomous agents, and recent "zero-copy" designs make different trade-offs whose impact on agentic workloads remains unclear. To clarify this space, we present BranchBench, a benchmark for evaluating branchable relational DBMSes under agentic exploration. We characterize five representative workloads-agentic software engineering, failure reproduction, data curation, MCTS, and simulation-and design parameterized macrobenchmarks that execute branch-mutate-evaluate loops to reflect these workloads, along with microbenchmarks that isolate branch lifecycle costs. We evaluate state of the art systems including Neon, DoltgreSQL, Tiger Data, Xata, and PostgreSQL baselines, and find a fundamental tension: systems optimized for fast branching suffer up to 5-4000x slower reads as branches deepen, while systems optimized for fast data operations incur 25-1500x higher branch creation and switching latency. Further, no current system supports the representative workloads at scale. These results highlight the need for branch-native DBMSes designed specifically for agentic exploration.

cs.DB↗

arXiv2026

Visual Decoding Operators: Towards a Compositional Theory of Visualization Perception

Prior work on perceptual effectiveness has decomposed visualizations into smaller common units (e.g., channels such as angle, position, and length) to establish rankings. While useful, these decompositions lack the computational structure to predict performance for new visualization x task combinations, requiring new experiments for each. We propose an alternative unit of analysis: operationalizing quantitative visualization interpretation as sequences of composable visual decoding operators. Using probability density function (PDF) and cumulative distribution function (CDF) charts, we examine how four chart-specific tasks can be decomposed into five reusable, chart-agnostic perceptual operations and characterize their error profiles through hierarchical Bayesian modeling. We then test generalizability by composing one kind of learned operators to predict performance on a structurally different task: Moritz et al.'s [37] scatterplot mean-estimation experiment, where the chart type, chart dimensions, and analytic goal all differ from the learning conditions. With a pre-registered analysis plan, we compose operators under six candidate strategies and evaluate each against empirical data with no parameters fit to the response data. One strategy captures both bias and variance of observed responses; five alternatives fail in distinguishable ways. We argue that this decoding-operator-oriented approach to empirical visualization research demonstrates the feasibility of a different way of doing empirical visualization research, one where findings compose, and predictions extend beyond the conditions in which they were measured. Free copy of this paper and supplemental materials: https://osf.io/prtfq.

cs.HC↗

arXiv2026

Human-Data Interaction, Exploration, and Visualization in the AI Era: Challenges and Opportunities

The rapid advancement of AI is transforming human-centered systems, with profound implications for human-AI interaction, human-data interaction, and visual analytics. In the AI era, data analysis increasingly involves large-scale, heterogeneous, and multimodal data that is predominantly unstructured, as well as foundation models such as LLMs and VLMs, which introduce additional uncertainty into analytical processes. These shifts expose persistent challenges for human-data interactive systems, including perceptually misaligned latency, scalability constraints, limitations of existing interaction and exploration paradigms, and growing uncertainty regarding the reliability and interpretability of AI-generated insights. Responding to these challenges requires moving beyond conventional efficiency and scalability metrics, redefining the roles of humans and machines in analytical workflows, and incorporating cognitive, perceptual, and design principles into every level of the human-data interaction stack. This paper investigates the challenges introduced by recent advances in AI and examines how these developments are reshaping the ways users engage with data, while outlining limitations and open research directions for building human-centered AI systems for interactive data analysis in the AI era.

cs.DB↗

arXiv2026

Outrunning LLM Cutoffs: A Live Kernel Crash Resolution Benchmark for All

Repairing system crashes discovered by kernel fuzzers like Syzkaller is a critical yet underexplored challenge in software engineering. While recent works have introduced Large Language Model (LLM) based agents for Linux kernel crash-resolution, their evaluation benchmarks are usually static and thus, do not capture the evolving nature of the Linux kernel, and suffer from potential data contamination due to LLM knowledge cutoffs. To address the above problem, we present (i) Live-kBench, an evaluation framework for self-evolving benchmarks that continuously scrapes and evaluates agents on freshly discovered kernel bugs, and (ii) kEnv, an agent-agnostic standardized crash-resolution environment for kernel compilation, execution, and feedback. This design decouples agent workflows from heavy-weight execution, enabling fair and scalable comparison across diverse agent frameworks under identical conditions. To this end, we curate an inaugural dataset of 534 Linux kernel bugs and empirically demonstrate a significant performance gap, with agents achieving up to 25% higher equivalent patch rate on bugs fixed before the LLM knowledge cutoff. Using kEnv, we benchmark three state-of-the-art agents, showing that they resolve 74% of crashes on the first attempt (plausible patches); however only ~20% of generated patches closely match developer fixes. Additionally, exposing crash resolution feedback improves crash resolution rate by 29%. Live-kBench provides the community with an evaluation infrastructure for self-evolving benchmarks that is both time and attribute sensitive; complete with a public dashboard to track agent progress on Linux kernel bugs.

cs.SE↗

arXiv2026

Data-Induced Groupings and How To Find Them

Making sense of a visualization requires the reader to consider both the visualization design and the underlying data values. Existing work in the visualization community has largely considered affordances driven by visualization design elements, such as color or chart type, but how visual design interacts with data values to impact interpretation and reasoning has remained under-explored. Dot plots and bar graphs are commonly used to help users identify groups of points that form trends and clusters, but are liable to manifest groupings that are artifacts of spatial arrangement rather than inherent patterns in the data itself. These ``Data-induced Groups'' can drive suboptimal data comparisons and potentially lead the user to incorrect conclusions. We conduct two user studies using dot plots as a case study to understand the prevalence of data-induced groupings. We find that users rely on data-induced groupings in both conditions despite the fact that trend-based groupings are irrelevant in nominal data. Based on the study results, we build a model to predict whether users are likely to perceive a given set of dot plot points as a group. We discuss two use cases illustrating how the model can assist visualization designers by both diagnosing potential user-perceived groupings in dot plots and offering redesigns that better accentuate desired groupings through data rearrangement.

cs.HC↗

arXiv2025

Please Don't Kill My Vibe: Empowering Agents with Data Flow Control

The promise of Large Language Model (LLM) agents is to perform complex, stateful tasks. This promise is stunted by significant risks - policy violations, process corruption, and security flaws - that stem from the lack of visibility and mechanisms to manage undesirable data flows produced by agent actions. Today, agent workflows are responsible for enforcing these policies in ad hoc ways. Just as data validation and access controls shifted from the application to the DBMS, freeing application developers from these concerns, we argue that systems should support Data Flow Controls (DFCs) and enforce DFC policies natively. This paper describes early work developing a portable instance of DFC for DBMSes and outlines a broader research agenda toward DFC for agent ecosystems.

cs.CR↗

arXiv2025

A Decade of Systems for Human Data Interaction

Human-data interaction (HDI) presents fundamentally different challenges from traditional data management. HDI systems must meet latency, correctness, and consistency needs that stem from usability rather than query semantics; failing to meet these expectations breaks the user experience. Moreover, interfaces and systems are tightly coupled; neither can easily be optimized in isolation, and effective solutions demand their co-design. This dependence also presents a research opportunity: rather than adapt systems to interface demands, systems innovations and database theory can also inspire new interaction and visualization designs. We survey a decade of our lab's work that embraces this coupling and argue that HDI systems are the foundation for reliable, interactive, AI-driven applications.

cs.DB↗

arXiv2025

An approach for systematic decomposition of complex llm tasks

Large Language Models (LLMs) suffer from reliability issues on complex tasks, as existing decomposition methods are heuristic and rely on agent or manual decomposition. This work introduces a novel, systematic decomposition framework that we call Analysis of CONstraint-Induced Complexity (ACONIC), which models the task as a constraint problem and leverages formal complexity measures to guide decomposition. On combinatorial (SAT-Bench) and LLM database querying tasks (Spider), we find that by decomposing the tasks following the measure of complexity, agent can perform considerably better.

cs.AI↗

arXiv2025

Toward Systems Foundations for Agentic Exploration

Agentic exploration, letting LLM-powered agents branch, backtrack, and search across many execution paths, demands systems support well beyond today's pass-at-k resets. Our benchmark of six snapshot/restore mechanisms shows that generic tools such as CRIU or container commits are not fast enough even in isolated testbeds, and they crumble entirely in real deployments where agents share files, sockets, and cloud APIs with other agents and human users. In this talk, we pinpoint three open fundamental challenges: fork semantics, which concerns how branches reveal or hide tentative updates; external side-effects, where fork awareness must be added to services or their calls intercepted; and native forking, which requires cloning databases and runtimes in microseconds without bulk copying.

cs.DC↗

arXiv2025

Steering Semantic Data Processing With DocWrangler

Unstructured text has long been difficult to automatically analyze at scale. Large language models (LLMs) now offer a way forward by enabling {\em semantic data processing}, where familiar data processing operators (e.g., map, reduce, filter) are powered by LLMs instead of code. However, building effective semantic data processing pipelines presents a departure from traditional data pipelines: users need to understand their data to write effective pipelines, yet they need to construct pipelines to extract the data necessary for that understanding -- all while navigating LLM idiosyncrasies and inconsistencies. We present \docwrangler, a mixed-initiative integrated development environment (IDE) for semantic data processing with three novel features to address the gaps between the user, their data, and their pipeline: {\em (i) In-Situ User Notes} that allows users to inspect, annotate, and track observations across documents and LLM outputs, {\em (ii) LLM-Assisted Prompt Refinement} that transforms user notes into improved operations, and {\em (iii) LLM-Assisted Operation Decomposition} that identifies when operations or documents are too complex for the LLM to correctly process and suggests decompositions. Our evaluation combines a think-aloud study with 10 participants and a public-facing deployment (available at \href{https://docetl.org/playground}{docetl.org/playground}) with 1,500+ recorded sessions, revealing how users develop systematic strategies for their semantic data processing tasks; e.g., transforming open-ended operations into classifiers for easier validation and intentionally using vague prompts to learn more about their data or LLM capabilities.

cs.HC↗

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

A Formalism and Library for Database Visualization

Existing data visualization formalisms are restricted to single-table inputs, which makes existing visualization grammars like Vega-lite or ggplot2 tedious to use, have overly complex APIs, and unsound when visualization multi-table data. This paper presents the first visualization formalism to support databases as input -- in other words, *database visualization*. A visualization specification is defined as a mapping from database constraints (e.g., schemas, types, foreign keys) to visual representations of those constraints, and we state that a visualization is *faithful* if it visually preserves the underlying database constraints. This formalism explains how visualization designs are the result of implicit data modeling decisions. We further develop a javascript library called dvl and use a series of case studies to show its expressiveness over specialized visualization systems and existing grammar-based languages.

cs.DB↗