Jinghang Gu
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
Research Subjects
Jinghang Gu has a total of 34 co-authors. Here are the top co-authors by number of shared publications:
- Chu-Ren Huang
The Hong Kong Polytechnic University
10 shared publications - Zhongqing Wang
School of Computer Science and Technology, Soochow University, Suzhou, China
7 shared publications Xiaoyi Bao
The Hong Kong Polytechnic University Hong Kong China
5 shared publications- Guodong Zhou
School of Computer Science and Technology, Soochow University, Suzhou, China
4 shared publications - Longhua Qian
Soochow University
4 shared publications - Pengchao Wu
Soochow University
3 shared publications
Latest Publications
- 15 September 2026
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 183
Unifying temporal knowledge Graph and LLMs for Biomedical Hypothesis generation via temporal soft Prompting
- 15 September 2026
- Published by Elsevier in Journal of Biomedical Informatics
- Vol. 183
Unifying temporal knowledge Graph and LLMs for Biomedical Hypothesis generation via temporal soft Prompting
- 31 August 2026
- Published by Wiley in International Journal of Applied Linguistics
How Is L2 Grit Positioned in Chinese College EFL Learners’ Network of Anxiety, Boredom, and Psychological Need Frustration? A Psychological Network Analysis
This study examines how domain‐specific L2 grit is conditionally associated with foreign language anxiety, foreign language boredom, and basic psychological need frustration among Chinese university EFL learners. Drawing on Complex Dynamic Systems Theory as a systems‐oriented heuristic, we use psychological network analysis to model these variables as an interconnected set of contemporaneous associations. Survey data were collected from 908 university students enrolled in English courses. Participants completed established scales measuring L2 grit, foreign language anxiety, classroom boredom, and psychological need frustration. To capture the interrelations among these constructs, we estimated a Gaussian graphical model in which factor scores derived from CFA‐estimated latent constructs were represented as nodes and regularized partial correlations were represented as edges. The estimated network showed relatively denser associations among boredom facets and psychological need‐frustration facets, whereas the two L2 grit facets displayed comparatively weak direct connections to the other nodes. These results indicate that, in this cross‐sectional network, L2 grit was not among the most structurally prominent nodes. The findings suggest that strengthening grit alone may be insufficient to explain variation in classroom anxiety, boredom, and psychological need frustration. Future intervention research should therefore examine whether emotion regulation support, task design, and psychological need support are more directly associated with improved classroom affect.
- 31 August 2026
- Published by Wiley in International Journal of Applied Linguistics
How Is L2 Grit Positioned in Chinese College EFL Learners’ Network of Anxiety, Boredom, and Psychological Need Frustration? A Psychological Network Analysis
This study examines how domain‐specific L2 grit is conditionally associated with foreign language anxiety, foreign language boredom, and basic psychological need frustration among Chinese university EFL learners. Drawing on Complex Dynamic Systems Theory as a systems‐oriented heuristic, we use psychological network analysis to model these variables as an interconnected set of contemporaneous associations. Survey data were collected from 908 university students enrolled in English courses. Participants completed established scales measuring L2 grit, foreign language anxiety, classroom boredom, and psychological need frustration. To capture the interrelations among these constructs, we estimated a Gaussian graphical model in which factor scores derived from CFA‐estimated latent constructs were represented as nodes and regularized partial correlations were represented as edges. The estimated network showed relatively denser associations among boredom facets and psychological need‐frustration facets, whereas the two L2 grit facets displayed comparatively weak direct connections to the other nodes. These results indicate that, in this cross‐sectional network, L2 grit was not among the most structurally prominent nodes. The findings suggest that strengthening grit alone may be insufficient to explain variation in classroom anxiety, boredom, and psychological need frustration. Future intervention research should therefore examine whether emotion regulation support, task design, and psychological need support are more directly associated with improved classroom affect.
- Open Access
Temporally Informed Distillation of Embedding Semantics: Beyond Continued Pretraining for Modeling Gender Ideology in Dated Texts
Modeling historically situated gender ideology remains challenging for language models, as contemporary embeddings struggle to reflect temporally specific semantic structures beyond surface lexical patterns. Although large language models exhibit extensive general-purpose performance, their direct use with history-specific semantic analysis is limited by the distributional mismatch between contemporary training data and historical linguistic patterns. These constraints encourage the distillation of temporally based semantic knowledge into small student architectures. To solve this issue, we propose Temporally Informed Distillation of Embedding Semantics (TIDES), which integrates continued pretraining on temporally specific corpora with feature-level distillation from large embedding teachers. We evaluate TIDES across teacher architectures with distinct pretraining objectives. While continued pretraining provides lexical and syntactic adaptation, our results show that improvements in ideological modeling cannot be attributed to additional training exposure alone. Rather, teacher–student structural alignment is also critical to transfer effectiveness. Contrastive, encoder-aligned teachers yield substantially more stable preservation of fine-grained, historically situated semantic distinctions. These findings suggest that temporal ideology transfer is representation-dependent: ideological meaning can be shaped by the geometry and training objectives of embedding spaces. By introducing TIDES and providing evidence that architectural compatibility can influence ideological inheritance, this study advances a representation-centered account of modeling ideology in temporally grounded semantic research.
Open AccessTemporally Informed Distillation of Embedding Semantics: Beyond Continued Pretraining for Modeling Gender Ideology in Dated Texts
Modeling historically situated gender ideology remains challenging for language models, as contemporary embeddings struggle to reflect temporally specific semantic structures beyond surface lexical patterns. Although large language models exhibit extensive general-purpose performance, their direct use with history-specific semantic analysis is limited by the distributional mismatch between contemporary training data and historical linguistic patterns. These constraints encourage the distillation of temporally based semantic knowledge into small student architectures. To solve this issue, we propose Temporally Informed Distillation of Embedding Semantics (TIDES), which integrates continued pretraining on temporally specific corpora with feature-level distillation from large embedding teachers. We evaluate TIDES across teacher architectures with distinct pretraining objectives. While continued pretraining provides lexical and syntactic adaptation, our results show that improvements in ideological modeling cannot be attributed to additional training exposure alone. Rather, teacher–student structural alignment is also critical to transfer effectiveness. Contrastive, encoder-aligned teachers yield substantially more stable preservation of fine-grained, historically situated semantic distinctions. These findings suggest that temporal ideology transfer is representation-dependent: ideological meaning can be shaped by the geometry and training objectives of embedding spaces. By introducing TIDES and providing evidence that architectural compatibility can influence ideological inheritance, this study advances a representation-centered account of modeling ideology in temporally grounded semantic research.
- 11 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 177
Sentimental image generation with image quality assessment
- 11 February 2026
- Published by Elsevier in Pattern Recognition
- Vol. 177
Sentimental image generation with image quality assessment
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38 (2) , 1070-1083
Exploring Context-Free Opinion Grammar for Aspect-Based Sentiment Analysis
Utilizing pre-trained generative models for sentiment element extraction has recently significantly enhanced aspect-based sentiment analysis benchmarks. Nonetheless, these models have two significant drawbacks: 1) high-computational cost in both the inference time and hardware requirement. 2) Lack of explicit modeling as they model the connections between sentiment elements with fragile natural or notational language target sequence. To overcome these challenges, we present a novel opinion tree parsing model designed to swiftly parse sentiment elements from an opinion tree. This approach not only accelerates the process but also explicitly unveils a more comprehensive and fully articulated aspect-level sentiment structure. Our method begins by introducing a pioneering context-free opinion grammar to standardize the opinion tree structure. Subsequently, we leverage a neural chart-based opinion tree parser to thoroughly explore the interconnections among sentiment elements and parse them into a structured opinion tree. Extensive experiments underscore the effectiveness of our proposed model and the capability of the opinion tree parser, particularly when coupled with the introduced context-free opinion grammar. Crucially, the results confirm the superior speed of our model compared to the SOTA baselines.
- 14 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Knowledge and Data Engineering
- Vol. 38 (2) , 1070-1083
Exploring Context-Free Opinion Grammar for Aspect-Based Sentiment Analysis
Utilizing pre-trained generative models for sentiment element extraction has recently significantly enhanced aspect-based sentiment analysis benchmarks. Nonetheless, these models have two significant drawbacks: 1) high-computational cost in both the inference time and hardware requirement. 2) Lack of explicit modeling as they model the connections between sentiment elements with fragile natural or notational language target sequence. To overcome these challenges, we present a novel opinion tree parsing model designed to swiftly parse sentiment elements from an opinion tree. This approach not only accelerates the process but also explicitly unveils a more comprehensive and fully articulated aspect-level sentiment structure. Our method begins by introducing a pioneering context-free opinion grammar to standardize the opinion tree structure. Subsequently, we leverage a neural chart-based opinion tree parser to thoroughly explore the interconnections among sentiment elements and parse them into a structured opinion tree. Extensive experiments underscore the effectiveness of our proposed model and the capability of the opinion tree parser, particularly when coupled with the introduced context-free opinion grammar. Crucially, the results confirm the superior speed of our model compared to the SOTA baselines.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4865-4877
CalligraphicOCR for Chinese Calligraphy Recognition
Xiaoyi Bao, Zhongqing Wang, Jinghang Gu, Chu-Ren Huang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4865-4877
Open AccessCalligraphicOCR for Chinese Calligraphy Recognition
Xiaoyi Bao, Zhongqing Wang, Jinghang Gu, Chu-Ren Huang. Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing. 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 8440-8451
Revisiting Classical Chinese Event Extraction with Ancient Literature Information
Xiaoyi Bao, Zhongqing Wang, Jinghang Gu, Chu-Ren Huang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 8440-8451
Open AccessRevisiting Classical Chinese Event Extraction with Ancient Literature Information
Xiaoyi Bao, Zhongqing Wang, Jinghang Gu, Chu-Ren Huang. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers). 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4070-4081
Sentimental Image Generation for Aspect-based Sentiment Analysis
Xiaoyi Bao, Jinghang Gu, Zhongqing Wang, Chu-Ren Huang. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4070-4081
Open AccessSentimental Image Generation for Aspect-based Sentiment Analysis
Xiaoyi Bao, Jinghang Gu, Zhongqing Wang, Chu-Ren Huang. Findings of the Association for Computational Linguistics: ACL 2025. 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 495-503
An Effective Incorporating Heterogeneous Knowledge Curriculum Learning for Sequence Labeling
Xuemei Tang, Jun Wang, Qi Su, Chu-Ren Huang, Jinghang Gu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 495-503
Open AccessAn Effective Incorporating Heterogeneous Knowledge Curriculum Learning for Sequence Labeling
Xuemei Tang, Jun Wang, Qi Su, Chu-Ren Huang, Jinghang Gu. Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers). 2025.
- Open Access
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4199-4210
Exploring Hybrid Sampling Inference for Aspect-based Sentiment Analysis
Xiaoyi Bao, Minjie Qiang, Jinghang Gu, Zhongqing Wang, Chu-Ren Huang. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.
- 1 January 2025
- Published by Association for Computational Linguistics (ACL)
- p. 4199-4210
Open AccessExploring Hybrid Sampling Inference for Aspect-based Sentiment Analysis
Xiaoyi Bao, Minjie Qiang, Jinghang Gu, Zhongqing Wang, Chu-Ren Huang. Findings of the Association for Computational Linguistics: NAACL 2025. 2025.