Selected work

Representative Papers

Open-Source Intelligence and Music Information Retrieval for Geographic Attribution of Musical Affect and the Ecological Limits of Population Inference

Jul 14, 2026

This study tests the hypothesis that regional music reflects population-level psychological traits, specifically examining whether musical emotional features can reliably infer national differences in well-being and individualism. Analyzing 2,393 unique melodies from the Essen Folksong Collection, the authors extracted eight melodic and affective features and correlated them with national indices from the World Happiness Report and Hofstede’s cultural dimensions. The work presents the first large-scale, systematic validation of measurable geographic signatures in symbolic music data, yet reveals a critical ecological fallacy: despite highly significant cross-national differences in melodic structure (p<0.001, with some effects as extreme as p<10⁻⁹⁰), none of the six tested musical-emotional indicators showed significant associations with national psychological metrics. The study establishes a reproducible analytical pipeline and proposes an empirical paradigm of “measurable but not inferable” for music-based psychological inference.

0 citationsRead paper

Open-Source Intelligence for Code Provenance and the Security Patterns that Separate Human and Large-Language-Model Implementations of Common Programming Tasks

Jul 14, 2026

This study investigates how to distinguish between human- and large language model (LLM)-generated code based solely on the code itself, with a focus on systematic differences in security-related programming behaviors. The authors introduce a reproducible, multilingual (Python, JavaScript, Go, Java) open-source analysis framework, assembling 528 samples across 31 security-sensitive tasks and 9 LLMs, evaluated through deterministic security detectors and cross-validated classifiers. Results show that a binary classifier achieves 93% accuracy (baseline: 78%) in identifying code provenance, while a seven-way classifier reaches 48%. Security pattern disparities are consistent across languages, yet code origin boundaries exhibit language-specific characteristics. Although LLMs successfully fix 77% of vulnerabilities, they frequently produce partial repairs—removing insecure patterns without implementing correct defenses. This work presents the first systematic characterization of behavioral differences in secure coding between humans and LLMs and proposes a data-driven, plug-and-play extensible analytical paradigm.

0 citationsRead paper

The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities

Jul 06, 2026

Current research on the security of execution environments for AI coding agents remains highly fragmented, lacking systematic integration and cross-disciplinary coordination. This work presents the first comprehensive survey of the field, analyzing 39 papers published between 2023 and 2026 and categorizing them into 17 thematic groups. Through CVE validation, cross-category comparison, and threat modeling, the study identifies critical disconnects among key areas such as isolation, access control, and time-of-check-to-time-of-use (TOCTOU) vulnerabilities, revealing five major research gaps. The analysis confirms four patched CVEs affecting production frameworks, quantifies the failure rate of existing mitigation strategies at 69%–98%, and uncovers that 17.1% of benign out-of-bound behaviors remain unaddressed by current mechanisms. Building on these findings, the paper proposes a unified research agenda to advance the field.

0 citationsRead paper

Unicode TAG-Block Concealment of Tool-Metadata Payloads in the Model Context Protocol: An Approval-View Fidelity Gap Across Three Independent Server Implementations

Jul 06, 2026

This work identifies and formally characterizes a fidelity gap in the Model Context Protocol (MCP) between the content displayed to human reviewers and what is actually ingested by language models, enabling attackers to inject stealthy malicious payloads. The authors propose a novel covert encoding mechanism leveraging Unicode TAG characters, which remain invisible under mainstream renderers yet are fully preserved by model tokenizers. Through a deterministic, protocol-level testing framework, they demonstrate the effectiveness of this approach across three independent MCP server implementations: all eight tested payloads successfully infiltrated the model context, with four bypassing conventional string-matching sanitizers. Notably, only TAG-based encoding simultaneously achieves human invisibility and full model retention, confirming that the vulnerability is inherent to the MCP specification itself.

0 citationsRead paper

Empirical Evaluation of Multi-Modal Touch Detection in Over-the-Shoulder Video Surveillance

Jun 28, 2026

This study addresses the challenge of reliably detecting and reconstructing touch inputs on mobile devices from over-the-shoulder videos under conditions where direct screen recording is unavailable. To this end, the authors propose a multimodal touch reconstruction framework that integrates hand anatomical keypoints (via MediaPipe), HSV-based skin-color segmentation, inter-frame difference for motion detection, and Canny edge analysis. The work presents the first systematic evaluation of such an approach in over-the-shoulder surveillance scenarios. Experimental results demonstrate that, in controlled settings, the method achieves an F1 score of 16.7% and a sequence similarity of 3.0%. However, in real-world uncontrolled videos, it suffers from severe false detections—averaging 57 spurious touch points per frame—rendering accurate keystroke sequence recovery impractical and highlighting a critical generalization bottleneck of current methods in real-world applications.

0 citationsRead paper

Recent publications

Latest Papers

Open-Source Intelligence and Music Information Retrieval for Geographic Attribution of Musical Affect and the Ecological Limits of Population Inference

Jul 14, 2026

This study tests the hypothesis that regional music reflects population-level psychological traits, specifically examining whether musical emotional features can reliably infer national differences in well-being and individualism. Analyzing 2,393 unique melodies from the Essen Folksong Collection, the authors extracted eight melodic and affective features and correlated them with national indices from the World Happiness Report and Hofstede’s cultural dimensions. The work presents the first large-scale, systematic validation of measurable geographic signatures in symbolic music data, yet reveals a critical ecological fallacy: despite highly significant cross-national differences in melodic structure (p<0.001, with some effects as extreme as p<10⁻⁹⁰), none of the six tested musical-emotional indicators showed significant associations with national psychological metrics. The study establishes a reproducible analytical pipeline and proposes an empirical paradigm of “measurable but not inferable” for music-based psychological inference.

0 citationsRead paper

Open-Source Intelligence for Code Provenance and the Security Patterns that Separate Human and Large-Language-Model Implementations of Common Programming Tasks

Jul 14, 2026

This study investigates how to distinguish between human- and large language model (LLM)-generated code based solely on the code itself, with a focus on systematic differences in security-related programming behaviors. The authors introduce a reproducible, multilingual (Python, JavaScript, Go, Java) open-source analysis framework, assembling 528 samples across 31 security-sensitive tasks and 9 LLMs, evaluated through deterministic security detectors and cross-validated classifiers. Results show that a binary classifier achieves 93% accuracy (baseline: 78%) in identifying code provenance, while a seven-way classifier reaches 48%. Security pattern disparities are consistent across languages, yet code origin boundaries exhibit language-specific characteristics. Although LLMs successfully fix 77% of vulnerabilities, they frequently produce partial repairs—removing insecure patterns without implementing correct defenses. This work presents the first systematic characterization of behavioral differences in secure coding between humans and LLMs and proposes a data-driven, plug-and-play extensible analytical paradigm.

0 citationsRead paper

The Balkanization of Execution-Security Research for AI Coding Agents: Isolation, Access Control, and Time-of-Check-to-Time-of-Use Vulnerabilities

Jul 06, 2026

Current research on the security of execution environments for AI coding agents remains highly fragmented, lacking systematic integration and cross-disciplinary coordination. This work presents the first comprehensive survey of the field, analyzing 39 papers published between 2023 and 2026 and categorizing them into 17 thematic groups. Through CVE validation, cross-category comparison, and threat modeling, the study identifies critical disconnects among key areas such as isolation, access control, and time-of-check-to-time-of-use (TOCTOU) vulnerabilities, revealing five major research gaps. The analysis confirms four patched CVEs affecting production frameworks, quantifies the failure rate of existing mitigation strategies at 69%–98%, and uncovers that 17.1% of benign out-of-bound behaviors remain unaddressed by current mechanisms. Building on these findings, the paper proposes a unified research agenda to advance the field.

0 citationsRead paper

Unicode TAG-Block Concealment of Tool-Metadata Payloads in the Model Context Protocol: An Approval-View Fidelity Gap Across Three Independent Server Implementations

Jul 06, 2026

This work identifies and formally characterizes a fidelity gap in the Model Context Protocol (MCP) between the content displayed to human reviewers and what is actually ingested by language models, enabling attackers to inject stealthy malicious payloads. The authors propose a novel covert encoding mechanism leveraging Unicode TAG characters, which remain invisible under mainstream renderers yet are fully preserved by model tokenizers. Through a deterministic, protocol-level testing framework, they demonstrate the effectiveness of this approach across three independent MCP server implementations: all eight tested payloads successfully infiltrated the model context, with four bypassing conventional string-matching sanitizers. Notably, only TAG-based encoding simultaneously achieves human invisibility and full model retention, confirming that the vulnerability is inherent to the MCP specification itself.

0 citationsRead paper

Empirical Evaluation of Multi-Modal Touch Detection in Over-the-Shoulder Video Surveillance

Jun 28, 2026

This study addresses the challenge of reliably detecting and reconstructing touch inputs on mobile devices from over-the-shoulder videos under conditions where direct screen recording is unavailable. To this end, the authors propose a multimodal touch reconstruction framework that integrates hand anatomical keypoints (via MediaPipe), HSV-based skin-color segmentation, inter-frame difference for motion detection, and Canny edge analysis. The work presents the first systematic evaluation of such an approach in over-the-shoulder surveillance scenarios. Experimental results demonstrate that, in controlled settings, the method achieves an F1 score of 16.7% and a sequence similarity of 3.0%. However, in real-world uncontrolled videos, it suffers from severe false detections—averaging 57 spurious touch points per frame—rendering accurate keystroke sequence recovery impractical and highlighting a critical generalization bottleneck of current methods in real-world applications.

0 citationsRead paper