arXiv subjects

Xi He

Publications and source records attributed to Xi He.

At least 19 recordsLinked to original sources

arXiv2026

ReBound: Reuse-Aware Privacy For Interactive Decision Support

Differentially private decision support frameworks answer complex aggregate threshold queries with formal bounds on false negative and false positive rates, but treat each query independently with no memory of past results. In practice, analysts work interactively, issuing sequences of related queries that refine bounds, adjust thresholds, or derive new functions from previous ones. We propose ReBound, a framework that reuses cached results from previous queries to answer new queries at reduced or zero additional privacy cost while maintaining formal utility guarantees. ReBound introduces a reuse framework for multiple refinement types, a cache graph structure for efficient lookup of reusable results, and a negotiation mechanism for when requested bounds cannot be met within budget.

cs.CR↗

arXiv2026

Traceable Knowledge Graph Reasoning Enables LLM-Assisted Decision Support for Industrial VOCs in the Steel Industry

Key knowledge for steel-industry volatile organic compounds (VOCs) governance is scattered across unstructured scientific literature, making it difficult to integrate process, pollutant, and control-technology evidence and increasing the risk of hallucination when general large language models (LLMs) answer low-frequency industrial questions. Here we developed Chat-ISV, a knowledge graph (KG) enhanced multi-agent Q&A system that parses a curated steel-industry VOCs literature corpus, constructs a Neo4j KG with 27180 nodes and 81779 semantic edges, and combines prompt-constrained extraction, chunk-centered topology optimization, multi-agent routing, source-backtracking retrieval, local literature retrieval, open-domain knowledge access, and interactive subgraph visualization. Benchmark tests and 400 expert blind evaluations showed that topology optimization reduced isolated nodes from 57% to 4.08% and that Chat-ISV achieved high factual reliability, with 96.93% precision, 72.63% recall, an F1-score of 0.830, and a mean score of 1.69/2.00. By converting fragmented environmental-engineering literature into traceable, queryable, and decision-support-oriented knowledge, Chat-ISV establishes a scalable environmental-informatics paradigm for reliable LLM deployment and intelligent pollution-control decision support in specialized industrial domains.

cs.AI↗

arXiv2026

FERMI: Exploiting Relations for Membership Inference Against Tabular Diffusion Models

Diffusion models are the leading approach for tabular data synthesis and are increasingly used to share sensitive records. Whether they actually protect privacy has become a pressing question. Membership inference attacks are the standard tool for this purpose, yet existing attacks assume a single-table setting and ignore the multi-relational structure of real sensitive data. A core challenge in assessing privacy risks from membership inference attacks in multi-table settings is how to leverage auxiliary information from relations associated with the target table, such as its parent tables. Particularly, we study a practical setting in which such auxiliary information is available only when training the attack model. At inference time, the attacker observes only the attribute values of the target record from the target table. We propose FERMI (FEature-mapping for Relational Membership Inference), which resolves this gap by enriching single-table features with relational membership signal. Across three tabular diffusion architectures and three real-world relational datasets, FERMI consistently improves attack performance over single-table baselines, with TPR@$0.1$FPR rising by up to 53% over the single-table baseline in the white-box setting and 22% in the black-box setting.

cs.LG↗

arXiv2026

On Privacy Leakage in Tabular Diffusion Models: Influential Factors, Attacker Knowledge, and Metrics

Tabular data plays an important role in many fields and industries, including those with elevated privacy considerations and risks. As such, there is a rising interest in generating high-quality synthetic proxies for real tabular data as a means of reducing privacy risk and proprietary data exposure. With tabular diffusion models (TDMs) demonstrating leading performance in synthesizing such data, understanding and measuring the privacy risks associated with these models is imperative. Leveraging state-of-the-art membership inference attacks for TDMs in both black- and white-box settings, this work quantifies the impact of training setup, synthesis choices, and attacker knowledge on privacy leakage. Moreover, the results demonstrate that adversaries need not have perfect knowledge of the training setup, identical data distributions, or massive compute resources to construct successful attacks. Finally, the pitfalls associated with applying heuristic privacy metrics, such as distance-to-closest record, are revealed.

cs.LG↗

arXiv2026

Variance Geometry of Exact Pauli-Detecting Codes: Continuous Landscapes Beyond Stabilizers

Exact quantum codes detecting a prescribed set of Pauli errors are approached through algebraic constructions--stabilizer, codeword-stabilized, permutation-invariant, topological, and related families. Geometrically, exact Pauli detection is governed by joint higher-rank numerical ranges of these Pauli operators, whose structure for rank $\geq 2$ is largely uncharted. From this viewpoint, we show that such codes often form connected continuous families rather than collections of disjoint solution regions. These families are characterized by a single scalar derived from the Knill-Laflamme conditions: denoted $\lambda^*$, it is the Euclidean norm of the signature vector of Pauli expectation values on the maximally mixed code state, and provides a one-parameter summary of the code's joint Pauli variance profile. Within these continuous landscapes, stabilizer codes occupy only discrete, measure-zero subsets of the attainable $\lambda^*$-spectrum, exposing a largely unexplored continuum of genuinely nonadditive exact codes. We establish this picture by analyzing the geometry of higher-rank operator compressions, and extend it to symmetry-restricted settings where cyclic and permutation symmetries are imposed on both the error model and the code projector. Small-system cases reveal interval, singleton, and empty regimes through eigenvalue interlacing and symmetry-sector decompositions; larger systems are treated numerically via Stiefel-manifold optimization and symmetry-adapted parameterizations. In every unrestricted and symmetry-compatible case analyzed, the attainable $\lambda^*$-spectrum forms a single closed interval whenever nonempty--although a general proof remains open. These results place stabilizer, symmetric, and nonadditive code families within a unified higher-rank variance framework, suggesting a continuous geometric perspective on the landscape of exact quantum codes.

quant-ph↗

arXiv2026

Interpreting the Error of Differentially Private Median Queries through Randomization Intervals

It can be difficult for practitioners to interpret the quality of differentially private (DP) statistics due to the added noise. One method to help analysts understand the amount of error introduced by DP is to return a Randomization Interval (RI), along with the statistic. A RI is a type of confidence interval that bounds the error introduced by DP. For queries where the noise distribution depends on the input, such as the median, prior work degrades the quality of the median itself to obtain a high-quality RI. In this work, we propose PostRI, a solution to compute a RI after the median has been estimated. PostRI enables a median estimation with 14%-850% higher utility than related work, while maintaining a narrow RI.

cs.CR↗

arXiv2026

MIDST Challenge at SaTML 2025: Membership Inference over Diffusion-models-based Synthetic Tabular data

Synthetic data is often perceived as a silver-bullet solution to data anonymization and privacy-preserving data publishing. Drawn from generative models like diffusion models, synthetic data is expected to preserve the statistical properties of the original dataset while remaining resilient to privacy attacks. Recent developments of diffusion models have been effective on a wide range of data types, but their privacy resilience, particularly for tabular formats, remains largely unexplored. MIDST challenge sought a quantitative evaluation of the privacy gain of synthetic tabular data generated by diffusion models, with a specific focus on its resistance to membership inference attacks (MIAs). Given the heterogeneity and complexity of tabular data, multiple target models were explored for MIAs, including diffusion models for single tables of mixed data types and multi-relational tables with interconnected constraints. MIDST inspired the development of novel black-box and white-box MIAs tailored to these target diffusion models as a key outcome, enabling a comprehensive evaluation of their privacy efficacy. The MIDST GitHub repository is available at https://github.com/VectorInstitute/MIDST

cs.LG↗

arXiv2026

CAPID: Context-Aware PII Detection for Question-Answering Systems

Detecting personally identifiable information (PII) in user queries is critical for ensuring privacy in question-answering systems. Current approaches mainly redact all PII, disregarding the fact that some of them may be contextually relevant to the user's question, resulting in a degradation of response quality. Large language models (LLMs) might be able to help determine which PII are relevant, but due to their closed source nature and lack of privacy guarantees, they are unsuitable for sensitive data processing. To achieve privacy-preserving PII detection, we propose CAPID, a practical approach that fine-tunes a locally owned small language model (SLM) that filters sensitive information before it is passed to LLMs for QA. However, existing datasets do not capture the context-dependent relevance of PII needed to train such a model effectively. To fill this gap, we propose a synthetic data generation pipeline that leverages LLMs to produce a diverse, domain-rich dataset spanning multiple PII types and relevance levels. Using this dataset, we fine-tune an SLM to detect PII spans, classify their types, and estimate contextual relevance. Our experiments show that relevance-aware PII detection with a fine-tuned SLM substantially outperforms existing baselines in span, relevance and type accuracy while preserving significantly higher downstream utility under anonymization.

cs.CR↗

arXiv2026

Observation of correlated plasmons in low-valence nickelates

The discovery of nickelate superconductors has opened a new arena for studying the behavior of correlated electron liquids that give rise to unconventional superconductivity. While critical information about a material's charge dynamics is encoded in its plasmons, collective modes of the electron gas, these excitations have not yet been observed in nickelate materials. Here, we use resonant inelastic x-ray scattering (RIXS) to detect plasmons in the metallic, low-valence nickelate Pr4Ni3O8. Although qualitatively similar to those in cuprates, the nickelate plasmons are more heavily damped and have a lower velocity than those in a cuprate at comparable doping, which we attribute to reduced electronic hopping and enhanced screening of the long-range Coulomb interactions. Furthermore, the plasmons in Pr4Ni3O8 soften with increasing temperature, in contrast to the cuprate, where plasmons remain at nearly fixed energy but become more strongly damped. Taken together, these results reveal a distinct charge-screening landscape in nickelates and place quantitative constraints on analogies to cuprates.

cond-mat.str-el↗

arXiv2025

Co-Designing Quantum Codes with Transversal Diagonal Gates via Multi-Agent Systems

Exact scientific discovery requires more than heuristic search: candidate constructions must be turned into exact objects and checked independently. We address this gap by extending TeXRA with an independent Lean 4 verification layer, turning it into a human-guided multi-agent platform for exact scientific discovery. The platform couples symbolic synthesis, combinatorial and linear-programming search, exact reconstruction of numerical candidates, and formal verification in Lean. We apply this platform to nonadditive quantum error-correcting codes with prescribed transversal diagonal gates within the subset-sum linear-programming (SSLP) framework. In the distance-2 regime where logical states occupy distinct residue classes, the platform yields a Lean-certified catalogue of 14,116 codes for $K\in\{2,3,4\}$ and up to six physical qubits, realizing cyclic logical orders 2 through 18, from which we extract closed-form infinite families. We also construct a residue-degenerate $((6,4,2))$ code implementing the logical controlled-phase gate $\mathrm{diag}(1,1,1,i)$. At distance 3, we resolve the transversal-$T$ problem for $((7,2,3))$ codes within the complementary binary-dihedral $\mathrm{BD}_{16}$ setting: among the 12 candidates surviving the SSLP filters, 10 admit exact realizations and 2 are excluded by no-go proofs. All accepted constructions, families, and no-go results are formalized and checked in Lean, illustrating how AI-assisted workflows can bridge search, exact reconstruction, and formal proof in the physical sciences.

quant-ph↗

arXiv2025

GPM: The Gaussian Pancake Mechanism for Planting Undetectable Backdoors in Differential Privacy

Differential privacy (DP) has become the gold standard for preserving individual privacy in data analysis. However, an implicit yet fundamental assumption underlying these rigorous privacy guarantees is the correct implementation and execution of DP mechanisms. Several incidents of unintended privacy loss have occurred due to numerical issues and inappropriate configurations of DP software, which have been successfully exploited in privacy attacks. To better understand the seriousness of defective DP software, we ask the following question: is it possible to elevate these passive defects into active privacy attacks while maintaining covertness? To address this question, we present the Gaussian pancake mechanism (GPM), a novel mechanism that is computationally indistinguishable from the widely used Gaussian mechanism (GM), yet exhibits arbitrarily weaker statistical DP guarantees. This unprecedented separation enables a new class of backdoor attacks: by indistinguishably passing off as the authentic GM, GPM can covertly degrade statistical privacy. Unlike the unintentional privacy loss caused by GM's numerical issues, GPM is an adversarial yet undetectable backdoor attack against data privacy. We formally prove GPM's covertness, characterize its statistical leakage, and demonstrate a concrete distinguishing attack that can achieve near-perfect success rates under suitable parameter choices, both theoretically and empirically. Our results underscore the importance of using transparent, open-source DP libraries and highlight the need for rigorous scrutiny and formal verification of DP implementations to prevent subtle, undetectable privacy compromises in real-world systems.

cs.CR↗

arXiv2025

MaskSQL: Safeguarding Privacy for LLM-Based Text-to-SQL via Abstraction

Large language models (LLMs) have shown promising performance on tasks that require reasoning, such as text-to-SQL, code generation, and debugging. However, regulatory frameworks with strict privacy requirements constrain their integration into sensitive systems. State-of-the-art LLMs are also proprietary, costly, and resource-intensive, making local deployment impractical. Consequently, utilizing such LLMs often requires sharing data with third-party providers, raising privacy concerns and risking noncompliance with regulations. Although fine-tuned small language models (SLMs) can outperform LLMs on certain tasks and be deployed locally to mitigate privacy concerns, they underperform on more complex tasks such as text-to-SQL translation. In this work, we introduce MaskSQL, a text-to-SQL framework that utilizes abstraction as a privacy protection mechanism to mask sensitive information in LLM prompts. Unlike redaction, which removes content entirely, or generalization, which broadens tokens, abstraction retains essential information while discarding unnecessary details, striking an effective privacy-utility balance for the text-to-SQL task. Moreover, by providing mechanisms to control the privacy-utility tradeoff, MaskSQL facilitates adoption across a broader range of use cases. Our experimental results show that MaskSQL outperforms leading SLM-based text-to-SQL models and achieves performance approaching state-of-the-art LLM-based models, while preserving privacy.

cs.CR↗

arXiv2025

Optimal hypersurface decision trees

The study of optimal decision trees has gained increasing attention in recent years; however, despite substantial progress, it still suffers from two major challenges: First, trees constructed by existing optimal decision tree (ODT) algorithms have limited expressivity, as they are typically restricted to axis-parallel splits or binary features. Second, these algorithms generally do not scale well to large datasets. These two challenges are intertwined: decision trees with more expressive splitting rules incur significantly higher combinatorial complexity, making the ODT problem even more difficult to solve when using complex splits. Building on He and Little's proper decision tree framework, we propose the first algorithm for solving the optimal hypersurface decision tree problem with time complexity $O\left(K!\times N^{DG+G}\right)$, where $G$ is a variable depends on both $K$ (tree size), $M$ (polynomial degree of hypersurface) and $D$ (data dimension). To the best of our knowledge, no known algorithm is capable of producing decision trees with hypersurface splits. Moreover, the proposed algorithm is inherently amenable to vectorization, enabling efficient parallelization. Its generic design pattern also allows it to be used to accelerate other ODT variants, such as axis-parallel decision trees. Furthermore, we identify an effective pruning strategy for the optimal hypersurface decision tree problem, which enables our algorithm to run significantly faster than the worst-case upper bound, together with an incremental procedure that reduces the cost of checking the feasibility of a single configuration from quadratic to linear time.

cs.LG↗

arXiv2025

Foundational theory for optimal decision tree problems. I. Algorithmic and geometric foundations

In the first paper (part I) of this series of two, we introduce four novel definitions of the ODT problems: three for size-constrained trees and one for depth-constrained trees. These definitions are stated unambiguously through executable recursive programs, satisfying all criteria we propose for a formal specification. In this sense, they resemble the "standard form" used in the study of general-purpose solvers. Grounded in algebraic programming theory-a relational formalism for deriving correct-by-construction algorithms from specifications-we can not only establish the existence or nonexistence of dynamic programming solutions but also derive them constructively whenever they exist. Consequently, the four generic problem definitions yield four novel optimal algorithms for ODT problems with arbitrary splitting rules that satisfy the axioms and objective functions of a given form. These algorithms encompass the known depth-constrained, axis-parallel ODT algorithm as the special case, while providing a unified, efficient, and elegant solution for the general ODT problem. In Part II, we present the first optimal hypersurface decision tree algorithm and provide comprehensive experiments against axis-parallel decision tree algorithms, including heuristic CART and state-of-the-art optimal methods. The results demonstrate the significant potential of decision trees with flexible splitting rules. Moreover, our framework is readily extendable to support algorithms for constructing even more flexible decision trees, including those with mixed splitting rules.

cs.LG↗

arXiv2025

Functional design of efficient and parallelizable combinatorial generators using convolution

The application of program transformation and algebraic methods to the development of efficient combinatorial optimization (CO) algorithms relies on an exhaustive combinatorial generator for the problem specification, followed by the fusion of thinning or filtering processes into this specification. However, the effectiveness of such fusion transformations critically depends on the structural compatibility between the objective function and the generator, which is highly problem dependent. In practice, when the majority of candidate solutions remain unfiltered or are not eliminated-as is the case for most intractable CO problems-the overall efficiency of the resulting fused program is largely determined by the intrinsic efficiency of the combinatorial generator. Consequently, if the specification itself exhibits suboptimal performance, the fused program will inherit a correspondingly inferior level of efficiency. We argue that a genuine designed process should also account for hardware compatibility and parallelizability-particularly the ability to support efficient parallel execution on modern hardware architectures, including multi-level cache hierarchies and GPUs. However, does achieving formal correctness necessarily conflict with designing algebraically elegant algorithms that support fusion? Can we obtain both simultaneously? In this paper, we show that techniques from functional programming, provide powerful formal tools for the systematic construction of such hardware-compatible and parallelizable combinatorial generators. This paper investigates generators for two of the most fundamental combinatorial structures-combinations and permutations-together with their natural extension to nested generators (e.g., combinations/permutations of combinations/permutations).

cs.DM↗

arXiv2025

Paper2Poster: Towards Multimodal Poster Automation from Scientific Papers

Academic poster generation is a crucial yet challenging task in scientific communication, requiring the compression of long-context interleaved documents into a single, visually coherent page. To address this challenge, we introduce the first benchmark and metric suite for poster generation, which pairs recent conference papers with author-designed posters and evaluates outputs on (i)Visual Quality-semantic alignment with human posters, (ii)Textual Coherence-language fluency, (iii)Holistic Assessment-six fine-grained aesthetic and informational criteria scored by a VLM-as-judge, and notably (iv)PaperQuiz-the poster's ability to convey core paper content as measured by VLMs answering generated quizzes. Building on this benchmark, we propose PosterAgent, a top-down, visual-in-the-loop multi-agent pipeline: the (a)Parser distills the paper into a structured asset library; the (b)Planner aligns text-visual pairs into a binary-tree layout that preserves reading order and spatial balance; and the (c)Painter-Commenter loop refines each panel by executing rendering code and using VLM feedback to eliminate overflow and ensure alignment. In our comprehensive evaluation, we find that GPT-4o outputs-though visually appealing at first glance-often exhibit noisy text and poor PaperQuiz scores, and we find that reader engagement is the primary aesthetic bottleneck, as human-designed posters rely largely on visual semantics to convey meaning. Our fully open-source variants (e.g. based on the Qwen-2.5 series) outperform existing 4o-driven multi-agent systems across nearly all metrics, while using 87% fewer tokens. It transforms a 22-page paper into a finalized yet editable .pptx poster - all for just $0.005. These findings chart clear directions for the next generation of fully automated poster-generation models. The code and datasets are available at https://github.com/Paper2Poster/Paper2Poster.

cs.CV↗

arXiv2025

Cybroc: Cyborgizing Broccoli for Longevity

Cybroc is a series of kinetic art installations exploring the recent proliferating populist longevity activism through the satirical cyborgization of broccoli. The artwork augments the symbol of health food-broccoli-with prosthetic limbs to perform so-called longevity-enhancing exercises such as cold plunges, treadmill running, brachiation (arm-swinging), sled pushing, etc.-all simulations of primal human survival tasks reframed as modern fitness routines. Despite its mechanical augmentations, the broccoli's inevitable decay and rotting after exhibiting high-intensity performances prompts reflection on the limits of biological enhancement and the ethics of human enhancement beyond natural capabilities, particularly transhumanist ideals. By juxtaposing a symbolic healthy vegetable with cutting-edge concepts of human enhancement, Cybroc challenges viewers to consider the intersection of nature, technology, and the human quest for extended lifespan in our transhuman era.

cs.CY↗

arXiv2025

Deep-ICE: the first globally optimal algorithm for minimizing 0-1 loss in two-layer ReLU and maxout networks

This paper introduces the first globally optimal algorithm for the empirical risk minimization problem of two-layer maxout and ReLU networks, i.e., minimizing the number of misclassifications. The algorithm has a worst-case time complexity of $O\left(N^{DK+1}\right)$, where $K$ denotes the number of hidden neurons and $D$ represents the number of features. It can be can be generalized to accommodate arbitrary computable loss functions without affecting its computational complexity. Our experiments demonstrate that the proposed algorithm provides provably exact solutions for small-scale datasets. To handle larger datasets, we introduce a novel coreset selection method that reduces the data size to a manageable scale, making it feasible for our algorithm. This extension enables efficient processing of large-scale datasets and achieves significantly improved performance, with a 20-30\% reduction in misclassifications for both training and prediction, compared to state-of-the-art approaches (neural networks trained using gradient descent and support vector machines), when applied to the same models (two-layer networks with fixed hidden nodes and linear models). The artifacts of the Deep-ICE algorithm can be found in https://github.com/XiHegrt/DeepICE-algorithm-artifacts.

cs.LG↗