Selected work

Representative Papers

Certificates in P and Subquadratic-Time Computation of Radius, Diameter, and all Eccentricities in Graphs

Mar 13, 2018

Computing graph radius, diameter, and all eccentricities is conjectured to require quadratic time under the Strong Exponential Time Hypothesis (SETH). Method: We introduce the notion of *node certificates*—compact auxiliary structures capturing local graph topology sufficient to infer global eccentricities—and establish a tight relationship between certificate size and graph probing complexity. Building on this, we design a randomized subquadratic algorithmic framework supporting queries from one-hop to all-pairs distances, analyzed via primal-dual techniques to yield the first parameterized theoretical guarantees for subquadratic eccentricity computation. Results: Empirical evaluation shows node certificates are significantly smaller than the graph size in real-world networks; our algorithm achieves tight subquadratic time complexity (e.g., $O(n^2 / log n)$) across diverse graph classes; it substantially improves practical runtime while enabling rigorous, provable performance bounds.

11 citationsRead paper

Ethical Risk Assessment of the Data Harnessing Process of LLM supported on Consensus of Well-known Multi-Ethical Frameworks

May 19, 2025Canadian AI

This work addresses the current lack of a systematic framework for evaluating ethical risks in data collection practices for large language models (LLMs). It proposes the first quantifiable assessment framework that integrates multiple prominent ethical theories, structuring evaluation around core ethical principles through a set of targeted questions and establishing a scoring system to measure ethical risk. This approach enables systematic, quantitative ethical auditing of LLM data curation processes. By offering a practical tool for assessing ethical compliance in AI development, the framework fills a critical gap in existing research—particularly in the integration of diverse ethical theories and the empirical evaluation of real-world data practices—thereby advancing the responsible development of artificial intelligence.

1 citationsRead paper

Quantum-Structured World Models (QSWMs) for Predictive Latent Dynamics

Aug 05, 2026

This work addresses the challenge of effectively modeling structured temporal dependencies in complex dynamical systems, which traditional world models struggle to capture. The authors propose Quantum-inspired State-space World Models (QSWMs), introducing for the first time complex-valued representations and density-matrix-like structures from quantum theory into latent state modeling. QSWMs integrate complex-valued neural networks, density-matrix-analogous latent variables, structured transition operators, and measurement-based decoding mechanisms to establish a novel inductive bias that ensures classical inclusiveness, predictive sufficiency, and structural compactness. Evaluated on elementary cellular automata tasks, QSWMs demonstrate superior local prediction performance, confirming their modeling potential, although challenges remain in long-horizon rollouts and variants leveraging full density matrix formalism.

0 citationsRead paper

Looking under the Wrong Lamppost: On the Limitations of Automated Translation Quality Estimation

Aug 04, 2026

Current automatic Machine Translation Quality Estimation (QE) systems lack reliability in real-world scenarios because their segment-level evaluations neglect critical dimensions such as discourse coherence, stylistic consistency, and rhetorical adequacy. Through theoretical analysis and empirical investigation, this study systematically uncovers structural limitations in QE—particularly concerning generalization capacity, data bias, overfitting, annotation noise, and the modeling of linguistic complexity—and, for the first time, identifies an inherent bottleneck rooted in the cognitive foundations of translation itself. These findings challenge the prevailing assumption that QE performance can be sufficiently improved merely by scaling up data or model capacity. The paper cautions against deploying existing QE systems as the sole basis for decision routing or bypassing human review in production environments and advocates redirecting future research toward automating human evaluation grounded in the Multidimensional Quality Metrics (MQM) framework.

0 citationsRead paper

Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss?

Aug 04, 2026

This work addresses the limitation of current optimizing compilers, which often miss profitable optimizations due to the absence of critical semantic information in program representations. To recover such overlooked semantics from heterogeneous C/C++ contexts and generate optimization code that preserves semantic contracts, the authors propose leveraging large language models (LLMs). They introduce SeGaBench, the first executable benchmark for semantics-driven compiler optimization, comprising 120 representative cases, and establish a novel paradigm wherein an LLM acts as a speculative semantic proposer, complemented by a semantic validator and a performance evaluation protocol. Experimental results demonstrate that the best-performing model produces semantically correct outputs in 94.8% of cases, achieves speedups of at least 1.05× in 83.3% of cases, and yields performance gains in 93.3% of the benchmark suite.

0 citationsRead paper

Recent publications

Latest Papers

Quantum-Structured World Models (QSWMs) for Predictive Latent Dynamics

Aug 05, 2026

This work addresses the challenge of effectively modeling structured temporal dependencies in complex dynamical systems, which traditional world models struggle to capture. The authors propose Quantum-inspired State-space World Models (QSWMs), introducing for the first time complex-valued representations and density-matrix-like structures from quantum theory into latent state modeling. QSWMs integrate complex-valued neural networks, density-matrix-analogous latent variables, structured transition operators, and measurement-based decoding mechanisms to establish a novel inductive bias that ensures classical inclusiveness, predictive sufficiency, and structural compactness. Evaluated on elementary cellular automata tasks, QSWMs demonstrate superior local prediction performance, confirming their modeling potential, although challenges remain in long-horizon rollouts and variants leveraging full density matrix formalism.

0 citationsRead paper

Looking under the Wrong Lamppost: On the Limitations of Automated Translation Quality Estimation

Aug 04, 2026

Current automatic Machine Translation Quality Estimation (QE) systems lack reliability in real-world scenarios because their segment-level evaluations neglect critical dimensions such as discourse coherence, stylistic consistency, and rhetorical adequacy. Through theoretical analysis and empirical investigation, this study systematically uncovers structural limitations in QE—particularly concerning generalization capacity, data bias, overfitting, annotation noise, and the modeling of linguistic complexity—and, for the first time, identifies an inherent bottleneck rooted in the cognitive foundations of translation itself. These findings challenge the prevailing assumption that QE performance can be sufficiently improved merely by scaling up data or model capacity. The paper cautions against deploying existing QE systems as the sole basis for decision routing or bypassing human review in production environments and advocates redirecting future research toward automating human evaluation grounded in the Multidimensional Quality Metrics (MQM) framework.

0 citationsRead paper

Can Large Language Models Recover Semantic Optimization Opportunities That Compilers Miss?

Aug 04, 2026

This work addresses the limitation of current optimizing compilers, which often miss profitable optimizations due to the absence of critical semantic information in program representations. To recover such overlooked semantics from heterogeneous C/C++ contexts and generate optimization code that preserves semantic contracts, the authors propose leveraging large language models (LLMs). They introduce SeGaBench, the first executable benchmark for semantics-driven compiler optimization, comprising 120 representative cases, and establish a novel paradigm wherein an LLM acts as a speculative semantic proposer, complemented by a semantic validator and a performance evaluation protocol. Experimental results demonstrate that the best-performing model produces semantically correct outputs in 94.8% of cases, achieves speedups of at least 1.05× in 83.3% of cases, and yields performance gains in 93.3% of the benchmark suite.

0 citationsRead paper

Rethinking Generative AI Literacy: An Integrative, Developmental, and Dialectical Framework for K-12 Teacher Education

Aug 03, 2026

This study addresses the growing misalignment between the rapid integration of generative artificial intelligence into classrooms and educators’ preparedness to use it responsibly. To bridge this gap, the authors propose the Responsible Artificial Intelligence Literacy in Education (RAIL-Ed) framework. Drawing on a systematic review of 67 studies and integrating perspectives from critical pedagogy, pragmatism, sociocultural theory, and human-centered traditions, RAIL-Ed uniquely positions ethics, equity, and agency as constitutive elements. The framework articulates an integrative, developmental, and dialectical model organized around six core pillars, accompanied by a three-stage maturity rubric. Designed to inform K–12 teacher education, curriculum design, and policy development, RAIL-Ed offers a theoretically grounded and empirically actionable foundation that aligns with AI literacy initiatives from UNESCO and the OECD/European Commission.

0 citationsRead paper

Schedule-Informed Temporal Fusion Forecasting of Hourly Airport Security-Checkpoint Throughput

Aug 03, 2026

This study addresses the challenge of hourly passenger flow forecasting for airport security staff scheduling by proposing a method that does not require explicit passenger–flight matching. To overcome the limitation that flight schedules typically provide only departure times, the authors innovatively employ a truncated Poisson kernel to map scheduled flights into an interpretable arrival intensity signal at security checkpoints. This signal is then integrated with historical throughput data, scheduled activities, and temporal features within a Temporal Fusion Transformer-based time series forecasting model. Experimental results demonstrate that the proposed approach achieves a weighted mean absolute percentage error (WMAPE) of 9.33% in direct six-hour-ahead predictions, outperforming RNN (12.16%) and LSTM (11.37%) baselines, with particularly strong performance during peak hours. Moreover, recursive forecasts over 24–96 hours maintain stable errors ranging from 10.60% to 11.04%.

0 citationsRead paper