Selected work

Representative Papers

A Syntax-Injected Approach for Faster and More Accurate Sentiment Analysis

Jun 21, 2024arXiv.org

In sentiment analysis (SA), conventional dependency parsing enhances accuracy and interpretability but incurs prohibitive computational overhead, hindering practical deployment. This paper proposes the Sequence Labeling Syntactic Parser (SELSP), the first approach to formulate dependency parsing as a lightweight sequence labeling task, enabling efficient syntactic integration. SELSP incorporates a polarity-aware sentiment lexicon, employs ternary and quinary classification schemes, and is rigorously evaluated via multi-model ablation studies. Compared to Stanza and VADER, SELSP achieves significant gains in both accuracy and inference speed; against Transformer-based baselines, it accelerates inference by multiple orders of magnitude while retaining competitive performance on ternary sentiment classification. Key contributions are: (i) pioneering a sequence labeling paradigm for dependency parsing; (ii) empirically validating that sentiment lexicons grounded in polarity discrimination differences yield superior performance; and (iii) achieving a balanced optimization of inference speed, predictive accuracy, and model interpretability.

1 citationsRead paper

Uplifting the Superpowers of Worst-Case-Optimal Join Algorithms

Aug 04, 2026

This work addresses the inefficiency of existing worst-case optimal (WCO) join algorithms in handling filtering conditions during graph query evaluation, which typically rely on costly pre- or post-processing steps. The paper introduces the first approach that natively integrates filtering operations into the WCO join framework by leveraging an extended Ring index structure. This design enables efficient support for complex pattern matching over property graphs and queries expressed in languages such as SPARQL and GQL, while preserving index compactness. Experimental results demonstrate that the proposed method significantly outperforms state-of-the-art baseline systems on filtered graph queries, with negligible additional storage overhead.

0 citationsRead paper

Bound-Founded Semantics for Answer Set Programming with Difference Constraints: Preliminary Report

Jul 23, 2026

This work addresses the lack of a unified logical semantic foundation in existing hybrid Answer Set Programming (ASP) solvers when handling linear constraints, which leads to inconsistent system behavior. It proposes a multi-sorted variant of Bound-founded Here-and-There logic (HTb), establishing for the first time a uniform semantic framework for ASP with difference constraints and formally defining the notion of foundedness for numeric variables. Grounded in equilibrium model theory and a semantic analysis of constraint atoms, the approach not only uncovers the root causes of semantic discrepancies among systems such as clingo[DL], clingcon, and flingo, but also provides a foundational platform capable of uniformly characterizing the semantics of diverse hybrid ASP systems. This framework further supports program simplification and facilitates the integration of future semantic principles.

0 citationsRead paper

User Preference Induction with LLMs for Offline Top-N Recommendation Evaluation

Jul 13, 2026

Offline Top-N recommendation evaluation is prone to distortion due to data sparsity and popularity bias, as it conventionally treats unobserved items as irrelevant. To address this limitation, this work proposes the first approach that leverages large language models (LLMs) to construct textual user preference profiles and employs these profiles to infer item relevance for candidate recommendations. By expanding the set of relevance labels beyond observed interactions, the method mitigates evaluation bias induced by sparse feedback, thereby enhancing the completeness, fairness, and robustness of offline evaluation protocols.

0 citationsRead paper

Quantum Dynamic Time Warping for Multivariate Time Series Classification

Jun 26, 2026

This work addresses the limitation of traditional dynamic time warping (DTW), which relies on Euclidean distance and struggles to capture cross-channel dependencies in multivariate time series. The authors propose a hybrid quantum DTW (qDTW) framework that replaces classical distance with a parameterized geometric structure in a quantum Hilbert space, enabling efficient alignment via dynamic programming. A key innovation is the introduction of a unified pre-embedding ansatz that decouples trainable entanglement from classical inputs, thereby avoiding phase ambiguity and information bottlenecks. The study also uncovers a trade-off governing spatiotemporal expressivity, offering design principles for multivariate quantum circuits. Experiments demonstrate that qDTW outperforms classical baselines on multivariate time series benchmarks up to eight dimensions, validating the high representational power of untrained quantum kernels and their efficacy in disentangling high-dimensional overlapping data.

0 citationsRead paper

Recent publications

Latest Papers

Uplifting the Superpowers of Worst-Case-Optimal Join Algorithms

Aug 04, 2026

This work addresses the inefficiency of existing worst-case optimal (WCO) join algorithms in handling filtering conditions during graph query evaluation, which typically rely on costly pre- or post-processing steps. The paper introduces the first approach that natively integrates filtering operations into the WCO join framework by leveraging an extended Ring index structure. This design enables efficient support for complex pattern matching over property graphs and queries expressed in languages such as SPARQL and GQL, while preserving index compactness. Experimental results demonstrate that the proposed method significantly outperforms state-of-the-art baseline systems on filtered graph queries, with negligible additional storage overhead.

0 citationsRead paper

Bound-Founded Semantics for Answer Set Programming with Difference Constraints: Preliminary Report

Jul 23, 2026

This work addresses the lack of a unified logical semantic foundation in existing hybrid Answer Set Programming (ASP) solvers when handling linear constraints, which leads to inconsistent system behavior. It proposes a multi-sorted variant of Bound-founded Here-and-There logic (HTb), establishing for the first time a uniform semantic framework for ASP with difference constraints and formally defining the notion of foundedness for numeric variables. Grounded in equilibrium model theory and a semantic analysis of constraint atoms, the approach not only uncovers the root causes of semantic discrepancies among systems such as clingo[DL], clingcon, and flingo, but also provides a foundational platform capable of uniformly characterizing the semantics of diverse hybrid ASP systems. This framework further supports program simplification and facilitates the integration of future semantic principles.

0 citationsRead paper

User Preference Induction with LLMs for Offline Top-N Recommendation Evaluation

Jul 13, 2026

Offline Top-N recommendation evaluation is prone to distortion due to data sparsity and popularity bias, as it conventionally treats unobserved items as irrelevant. To address this limitation, this work proposes the first approach that leverages large language models (LLMs) to construct textual user preference profiles and employs these profiles to infer item relevance for candidate recommendations. By expanding the set of relevance labels beyond observed interactions, the method mitigates evaluation bias induced by sparse feedback, thereby enhancing the completeness, fairness, and robustness of offline evaluation protocols.

0 citationsRead paper

Quantum Dynamic Time Warping for Multivariate Time Series Classification

Jun 26, 2026

This work addresses the limitation of traditional dynamic time warping (DTW), which relies on Euclidean distance and struggles to capture cross-channel dependencies in multivariate time series. The authors propose a hybrid quantum DTW (qDTW) framework that replaces classical distance with a parameterized geometric structure in a quantum Hilbert space, enabling efficient alignment via dynamic programming. A key innovation is the introduction of a unified pre-embedding ansatz that decouples trainable entanglement from classical inputs, thereby avoiding phase ambiguity and information bottlenecks. The study also uncovers a trade-off governing spatiotemporal expressivity, offering design principles for multivariate quantum circuits. Experiments demonstrate that qDTW outperforms classical baselines on multivariate time series benchmarks up to eight dimensions, validating the high representational power of untrained quantum kernels and their efficacy in disentangling high-dimensional overlapping data.

0 citationsRead paper

kNN estimation in semi-functional partial linear regression with missing responses at random

Jun 18, 2026

This study addresses the semi-functional partially linear regression model with both finite-dimensional and functional covariates under randomly missing response data. It introduces, for the first time, the k-nearest neighbors (kNN) approach into this framework and proposes three kNN-based estimation strategies for the simultaneous estimation of finite-dimensional parameters and infinite-dimensional nonparametric operators. The proposed methodology offers a novel nonparametric modeling perspective for handling mixed-dimensional covariates with missing responses. Theoretically, the asymptotic consistency of the proposed estimators is rigorously established, demonstrating their feasibility and effectiveness in practical applications.

0 citationsRead paper