Generative Models Enhanced by Sequence Labelling and Aspect-Code Switching Improve Cross-lingual Aspect-Based Sentiment Analysis
本文提出SeqLab框架，结合序列标注和方面代码切换技术，以增强跨语言基于方面的情感分析，提升多情感元素任务的表现。
0 citationsRead paper
Selected work
Representative Papers
Novels generated by language models show compressed formal variation
This study investigates whether large language models can replicate the formal diversity of human-authored fiction at syntactic, readability, and lexical levels. Analyzing six corpora—including Victorian-era and zero-style novels generated by GPT-5.5 and Qwen3-14B alongside their human-written counterparts—the work introduces the novel concepts of “variance closure” and “correlation closure,” employing quantitative metrics such as MATTR-500, Shannon entropy, mean sentence length, readability indices, and punctuation usage rates. The findings reveal that while individual AI-generated texts approximate human stylistic patterns, they exhibit significantly compressed overall formal variation: syntactic structures, sentence lengths, readability profiles, and punctuation diversity are markedly less varied, lexical richness is constrained, and stable cross-metric correlation patterns characteristic of human writing are absent.
0 citationsRead paper
TinyML for Acoustic Anomaly Detection in IoT Sensor Networks
This work proposes a lightweight TinyML-based anomaly sound detection system tailored for microcontrollers to address the latency, high energy consumption, and privacy concerns associated with cloud-centric processing in IoT acoustic sensing. By extracting Mel-frequency cepstral coefficients (MFCCs) directly on the edge device and deploying a compressed and optimized neural network classifier, the system achieves high-accuracy, low-power, real-time local anomaly detection for the first time. Evaluated on the UrbanSound8K dataset, the approach attains 91% accuracy and a balanced F1-score of 0.91, demonstrating a practical trade-off between privacy preservation, energy efficiency, and real-time performance. This solution offers a viable pathway for scalable deployment in resource-constrained IoT applications.
0 citationsRead paper
Extending Czech Aspect-Based Sentiment Analysis with Opinion Terms: Dataset and LLM Benchmarks
This study addresses the scarcity of aspect-based sentiment analysis (ABSA) resources for low-resource languages like Czech, particularly the lack of datasets annotated with opinion terms and effective cross-lingual transfer methods. To bridge this gap, the authors construct the first Czech ABSA benchmark dataset in the restaurant domain, featuring opinion-term annotations and supporting three levels of task complexity. They systematically evaluate a range of Transformer and large language models (LLMs) under monolingual, cross-lingual, and multilingual settings, and propose an LLM-driven translation–label alignment strategy to enhance cross-lingual transfer. Experimental results demonstrate that the proposed approach significantly outperforms baseline methods on Czech ABSA, while also revealing limitations of current models in capturing fine-grained opinion expressions, thereby establishing a new benchmark for sentiment analysis in low-resource languages.
0 citationsRead paper
On- and Off-Chain Demand and Supply Drivers of Bitcoin Price
This study addresses a critical gap in the literature by systematically integrating daily on-chain and off-chain supply and demand data from 2019 to 2024 to examine their causal effects on Bitcoin prices, using an autoregressive distributed lag (ARDL) model. While existing research predominantly focuses on on-chain transactions, this work highlights that off-chain activity—accounting for approximately three-quarters of all Bitcoin transactions—plays a pivotal role in price dynamics. The findings reveal that off-chain demand exerts a significant long-run positive effect on Bitcoin prices, whereas only on-chain demand demonstrates long-term explanatory power among on-chain variables. Although whale transactions exhibit notable short-term impacts, their influence dissipates over the long run. These results uncover a “dual-driver” mechanism underlying Bitcoin price formation, offering a novel perspective on the pricing of crypto-assets.
0 citationsRead paper
Recent publications
Latest Papers
Generative Models Enhanced by Sequence Labelling and Aspect-Code Switching Improve Cross-lingual Aspect-Based Sentiment Analysis
本文提出SeqLab框架，结合序列标注和方面代码切换技术，以增强跨语言基于方面的情感分析，提升多情感元素任务的表现。
0 citationsRead paper
Novels generated by language models show compressed formal variation
This study investigates whether large language models can replicate the formal diversity of human-authored fiction at syntactic, readability, and lexical levels. Analyzing six corpora—including Victorian-era and zero-style novels generated by GPT-5.5 and Qwen3-14B alongside their human-written counterparts—the work introduces the novel concepts of “variance closure” and “correlation closure,” employing quantitative metrics such as MATTR-500, Shannon entropy, mean sentence length, readability indices, and punctuation usage rates. The findings reveal that while individual AI-generated texts approximate human stylistic patterns, they exhibit significantly compressed overall formal variation: syntactic structures, sentence lengths, readability profiles, and punctuation diversity are markedly less varied, lexical richness is constrained, and stable cross-metric correlation patterns characteristic of human writing are absent.
0 citationsRead paper
TinyML for Acoustic Anomaly Detection in IoT Sensor Networks
This work proposes a lightweight TinyML-based anomaly sound detection system tailored for microcontrollers to address the latency, high energy consumption, and privacy concerns associated with cloud-centric processing in IoT acoustic sensing. By extracting Mel-frequency cepstral coefficients (MFCCs) directly on the edge device and deploying a compressed and optimized neural network classifier, the system achieves high-accuracy, low-power, real-time local anomaly detection for the first time. Evaluated on the UrbanSound8K dataset, the approach attains 91% accuracy and a balanced F1-score of 0.91, demonstrating a practical trade-off between privacy preservation, energy efficiency, and real-time performance. This solution offers a viable pathway for scalable deployment in resource-constrained IoT applications.
0 citationsRead paper
Extending Czech Aspect-Based Sentiment Analysis with Opinion Terms: Dataset and LLM Benchmarks
This study addresses the scarcity of aspect-based sentiment analysis (ABSA) resources for low-resource languages like Czech, particularly the lack of datasets annotated with opinion terms and effective cross-lingual transfer methods. To bridge this gap, the authors construct the first Czech ABSA benchmark dataset in the restaurant domain, featuring opinion-term annotations and supporting three levels of task complexity. They systematically evaluate a range of Transformer and large language models (LLMs) under monolingual, cross-lingual, and multilingual settings, and propose an LLM-driven translation–label alignment strategy to enhance cross-lingual transfer. Experimental results demonstrate that the proposed approach significantly outperforms baseline methods on Czech ABSA, while also revealing limitations of current models in capturing fine-grained opinion expressions, thereby establishing a new benchmark for sentiment analysis in low-resource languages.
0 citationsRead paper
On- and Off-Chain Demand and Supply Drivers of Bitcoin Price
This study addresses a critical gap in the literature by systematically integrating daily on-chain and off-chain supply and demand data from 2019 to 2024 to examine their causal effects on Bitcoin prices, using an autoregressive distributed lag (ARDL) model. While existing research predominantly focuses on on-chain transactions, this work highlights that off-chain activity—accounting for approximately three-quarters of all Bitcoin transactions—plays a pivotal role in price dynamics. The findings reveal that off-chain demand exerts a significant long-run positive effect on Bitcoin prices, whereas only on-chain demand demonstrates long-term explanatory power among on-chain variables. Although whale transactions exhibit notable short-term impacts, their influence dissipates over the long run. These results uncover a “dual-driver” mechanism underlying Bitcoin price formation, offering a novel perspective on the pricing of crypto-assets.
0 citationsRead paper