Current Organizations / Institutes

The organizations/institutes from the scholar's latest article

Previous Organizations / Institutes

The organizations / institutes from the scholar's other articles in the past ten years


Longhui Zou has a total of 16 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Using Translog-II for Conducting Keylogging Experiments

    by Longhui Zou, Michael Carl

    Translation process research (TPR) relies on objective behavioral data to uncover the cognitive mechanisms underlying translation. This paper provides a comprehensive methodology for using Translog-II, a specialized tool for recording user activity data during translation tasks, integrated with the CRITT Translation Process Research Database (TPR-DB) for advanced post-processing. We outline the complete experimental workflow—from project configuration and data collection to word alignment, error annotation, and data visualization—demonstrated through an English-to-Chinese translation-from-scratch case study. A central contribution of the paper is a systematic analysis of the boundary mismatch between Input Method Editor (IME) conversion units and tokenized words in Chinese translation data. Using 62 from-scratch English-to-Chinese translation sessions from the RUC17 study, we identify three types of IME-tokenization relationships, quantify their frequency, and assess their impact on the reliability of alignment-group process measures. These findings, together with the documented workflow, establish a reproducible framework for keystroke-based translation process research that enhances the precision of data-driven insights into cognitive translation processes, particularly for languages that depend on IME-mediated input.

  • Open Access

    Using Translog-II for Conducting Keylogging Experiments

    by Longhui Zou, Michael Carl

    Translation process research (TPR) relies on objective behavioral data to uncover the cognitive mechanisms underlying translation. This paper provides a comprehensive methodology for using Translog-II, a specialized tool for recording user activity data during translation tasks. We outline the complete experimental workflow—from project configuration to data collection—demonstrated through an English-to-Chinese translation-from-scratch case study. The study details the integration of Translog-II with the CRITT Translation Process Research Database (TPR-DB) to facilitate advanced post-processing. Key technical challenges are addressed, specifically the complexities of keystroke-to-word mapping for logo-graphic scripts requiring Input Method Editors (IMEs). We further demonstrate automated alignment protocols, multidimensional error annotation, and data visualization techniques utilizing Python scripts and Shiny R interfaces. The results indicate that while automated mapping is generally robust, specific technical noise, particularly regarding long deletions, can be mitigated through systematic analysis. Ultimately, this protocol establishes a reproducible framework for exploring translator behavior, enhancing the precision of data-driven insights into cognitive translation processes.

  • How dependency-based syntactic complexity shapes post-editing of LLM-generated translations

    by Longhui Zou, Michael Carl, Jia Feng

    This study examines how dependency-based syntactic complexity shapes English-to-Chinese post-editing of GPT-4 output. We address three questions: (RQ1) whether syntactic complexity on the source-text and GPT-output sides predicts post-editing quality; (RQ2) whether translator expertise moderates syntactic complexity effects; and (RQ3) whether task conditions affect performance and moderate syntactic complexity effects. We collected data from 46 participants (30 students, 16 professionals) post-editing in Trados Studio. We operationalized complexity with dependency-based metrics grounded in Incomplete Dependency Theory and Dependency Locality Theory, complemented by English–Chinese–specific indices (Left-Embeddedness; Nested Noun Distance). Results show that complexity in both source texts and GPT-4 output predicts post-editing errors (RQ1); experts are less sensitive to rising complexity than students, especially at higher complexity levels (RQ2); termbase access reduces overall and terminology errors, and the expert advantage in handling complexity is larger under Light post-editing but attenuated under Full post-editing (RQ3).


  • Open Access

    Beyond automated metrics: Assessing GPT-4o and Google Translate against professional translation standards

    by Longhui Zou, Ali Saeedi, Geoffrey S. Koby

    This study evaluates the translation capabilities of GPT-4o, a large language model (LLM), and Google Translate, a neural machine translation (NMT) system, using the American Translators Association (ATA) certification examination framework. We assess translations in two high-resource language pairs: English-to-Chinese (eng-chi) and English-to-Arabic (eng-ara). The evaluation combines both automatic metrics using COMET and manual assessment by ATA-certified graders following the standardized ATA grading framework. Two source texts from retired ATA certification exams were translated by both systems, producing eight target texts in total. Our findings indicate varying performance across systems and language pairs, with only GPT-4o’s eng-ara translations achieving superior quality for both required texts. Error analysis reveals distinct patterns between systems and language pairs: GPT-4o’s eng-chi translations primarily exhibit challenges with Terminology, Literalness, and Omission, while Google Translate shows a different distribution dominated by Cohesion issues, followed by Literalness and Misunderstanding. For eng-ara translations, both systems display similar error patterns, primarily in Terminology and Literalness, suggesting consistent challenges in this language pair. While COMET scores indicate high performance across all translations, manual assessment reveals more nuanced distinctions in translation quality, particularly in handling rhetorical expressions and idiomatic language use. These findings highlight the importance of complementing automatic metrics with human assessment in translation quality evaluation. The study also suggests that translation challenges extend beyond text complexity, reflecting distinct linguistic characteristics of each language pair and varying approaches in handling these challenges by different machine translation (MT) systems.


  • Open Access

    Patterns of Attention and Quality in English-Chinese Simultaneous Interpreting with Text

    by Longhui Zou, Michael Carl, Jia Feng

    Simultaneous interpreting with text (SIMTXT) is generally seen as being cognitively more demanding than either simultaneous interpreting (SI) or sight translation (STT). However, little research has been done on how interpreters allocate their attention proportionally during SIMTXT between their auditory and visual attention. This study examines attention patterns during SIMTXT and investigates the relationship between the interpreters’ attention patterns and the quality of the interpreting output. Nine professional interpreters were recruited for SIMTXT, interpreting six English STs into Chinese. The interpreters listened to the audio input while the transcribed text of the audio input was displayed on the screen by Translog-II. The eye movements were recorded by an eye-tracker (Tobii TX300) while their interpretations were recorded by Audacity. The gaze and the spoken data (source and target) were synchronized on a word-level and aligned with the final STs and TTs. We explored and categorized the visual and auditory attention patterns based on their ear-voice span (EVS), eye-voice span (IVS), and ear-eye span (EIS) and found three types of attention patterns in our data, namely, ear-dominant (ED), eye-dominant (ID), and ear-eye-balanced (EIB). In addition, the interpreting output was annotated by three professional interpreters based on an error taxonomy adapted from Multidimensional Quality Metrics (MQM). We found that the EIB interpreters produced the lowest translation quality in terms of total error numbers and accuracy, followed by ED, and then ID interpreters. ED interpreters produced the highest translation quality in terms of fluency, followed by EIB, and then ID interpreters.


  • Open Access

    How Does Manufacturing Agglomeration Affect Green Development? A Spatial and Nonlinear Perspective

    by Huaxi Yuan, Longhui Zou, Xiangyong Luo, Yidai Feng

    Developing high-quality manufacturing industries and realizing green transformation are relatively pressing issues in the 21st century. Existing studies only focus on the economic or environmental effects of agglomeration, and the combined effects of manufacturing agglomeration have been neglected. Therefore, by referring to industrial agglomeration theory and constructing a theoretical analytical framework for manufacturing agglomeration and green development, this paper adopts the spatial panel Durbin model and mediating effect model with the panel data from China’s Yangtze River Economic Belt to empirically test the influence and its mechanism of manufacturing agglomeration on green development. The results show that: (1) There are significant temporal and spatial differences in green development in the Yangtze River Economic Belt. Overall, green development has maintained a steady increase on the timeline, but each region shows a hierarchical structure of “multiple peaks-multiple centers”. (2) There is a typical inverted U-shaped relationship between manufacturing agglomeration and green development, and the linear and quadratic coefficients of manufacturing agglomeration are −0.585 and −0.167, respectively. (3) Under the constraints of temporal, spatial, and urban heterogeneity, the impacts of manufacturing agglomeration on green development show significant differences. (4) Manufacturing agglomeration affects green development through three paths: the labor force upgrading effect, industrial structure upgrading effect, and technological innovation effect. The study can provide a theoretical and empirical basis for the green development of developing countries around the world.


  • Does manufacturing agglomeration promote or hinder green development efficiency? Evidence from Yangtze River Economic Belt, China

    by Huaxi Yuan, Longhui Zou, Yidai Feng, Lei Huang

    Sustainable development can be mainly achieved by promoting the green transformation and development of the world economy and by improving the efficiency of regional green development, which often receive extensive attention from the academia. This paper uses a spatial econometric model to estimate the impact of manufacturing agglomeration on green development efficiency based on the panel data of China’s Yangtze River Economic Belt (YREB). The results show an overall large gap of green development efficiency between regions in the Yangtze River Economic Zone, mostly due to the extremely uneven development of green development efficiency in the upper reaches. Opposite to the middle and lower reaches, manufacturing agglomeration in the upper reaches of the YREB improves green development efficiency. Manufacturing agglomeration is conducive to the improvement of green development efficiency in adjacent areas. Nonetheless, it may hinder green development efficiency by inhibiting green technological innovation. This paper provides empirical evidence and policy implications for applying manufacturing agglomeration to promote green development efficiency in accordance with local conditions.


  • AI-Based Syntactic Complexity Metrics and Sight Interpreting Performance

    by Longhui Zou, Michael Carl, Mehdi Mirzapour, Hélène Jacquenet, Lucas Nunes Vieira