{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,28]],"date-time":"2026-04-28T19:30:25Z","timestamp":1777404625769,"version":"3.51.4"},"reference-count":33,"publisher":"MDPI AG","issue":"8","license":[{"start":{"date-parts":[[2025,8,19]],"date-time":"2025-08-19T00:00:00Z","timestamp":1755561600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["61207429"],"award-info":[{"award-number":["61207429"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Computers"],"abstract":"<jats:p>With the growing demand for multilingual financial information, cross-lingual financial sentiment recognition faces significant challenges, including semantic misalignment, ambiguous sentiment expression, and insufficient transferability. To address these issues, a unified multilingual recognition framework is proposed, integrating semantic contrastive learning with a language-adaptive modulation mechanism. This approach is built upon the XLM-R multilingual model and employs a semantic contrastive module to enhance cross-lingual semantic consistency. In addition, a language modulation module based on low-rank parameter injection is introduced to improve the model\u2019s sensitivity to fine-grained emotional features in low-resource languages such as Chinese and French. Experiments were conducted on a constructed trilingual financial sentiment dataset encompassing English, Chinese, and French. The results demonstrate that the proposed model significantly outperforms existing methods in cross-lingual sentiment recognition tasks. Specifically, in the English-to-French transfer setting, the model achieved 73.6% in accuracy, 69.8% in F1-Macro, 72.4% in F1-Weighted, and a cross-lingual generalization score of 0.654. Further improvements were observed under multilingual joint training, reaching 77.3%, 73.6%, 76.1%, and 0.696, respectively. In overall comparisons, the proposed model attained the highest performance across cross-lingual scenarios, with 75.8% in accuracy, 72.3% in F1-Macro, and 74.7% in F1-Weighted, surpassing strong baselines such as XLM-R+SimCSE and LaBSE. These results highlight the model\u2019s superior capability in semantic alignment and generalization across languages. The proposed framework demonstrates strong applicability and promising potential in multilingual financial sentiment analysis, public opinion monitoring, and multilingual risk modeling.<\/jats:p>","DOI":"10.3390\/computers14080338","type":"journal-article","created":{"date-parts":[[2025,8,19]],"date-time":"2025-08-19T07:56:06Z","timestamp":1755590166000},"page":"338","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":2,"title":["Leveraging Contrastive Semantics and Language Adaptation for Robust Financial Text Classification Across Languages"],"prefix":"10.3390","volume":"14","author":[{"given":"Liman","family":"Zhang","sequence":"first","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qianye","family":"Lin","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fanyu","family":"Meng","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Siyu","family":"Liang","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingxuan","family":"Lu","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shen","family":"Liu","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kehan","family":"Chen","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yan","family":"Zhan","sequence":"additional","affiliation":[{"name":"National School of Development, Peking University, Beijing 100871, China"},{"name":"Artificial Intelligence Research Institute, Tsinghua University, Beijing 100084, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2025,8,19]]},"reference":[{"key":"ref_1","doi-asserted-by":"crossref","first-page":"45","DOI":"10.1007\/s13278-025-01461-8","article-title":"Generalizing sentiment analysis: A review of progress, challenges, and emerging directions","volume":"15","author":"Alahmadi","year":"2025","journal-title":"Soc. Netw. Anal. Min."},{"key":"ref_2","doi-asserted-by":"crossref","first-page":"181","DOI":"10.1016\/j.ins.2014.04.034","article-title":"Stream-based active learning for sentiment analysis in the financial domain","volume":"285","year":"2014","journal-title":"Inf. Sci."},{"key":"ref_3","doi-asserted-by":"crossref","first-page":"21463","DOI":"10.1007\/s00521-023-08470-8","article-title":"Speech emotion recognition and text sentiment analysis for financial distress prediction","volume":"35","author":"Hajek","year":"2023","journal-title":"Neural Comput. Appl."},{"key":"ref_4","doi-asserted-by":"crossref","first-page":"723","DOI":"10.1007\/s11280-013-0221-9","article-title":"Building emotional dictionary for sentiment analysis of online news","volume":"17","author":"Rao","year":"2014","journal-title":"World Wide Web"},{"key":"ref_5","unstructured":"Yuan, C., Liu, Y., Yin, R., Zhang, J., Zhu, Q., Mao, R., and Xu, R. (2020, January 13\u201315). Target-based sentiment annotation in Chinese financial news. Proceedings of the Twelfth Language Resources and Evaluation Conference, Marseille, France."},{"key":"ref_6","doi-asserted-by":"crossref","first-page":"3","DOI":"10.1186\/s40537-017-0111-6","article-title":"Big Data: Deep Learning for financial sentiment analysis","volume":"5","author":"Sohangir","year":"2018","journal-title":"J. Big Data"},{"key":"ref_7","first-page":"324","article-title":"Sentiment Analysis and Emotion Detection Using Transformer Models in Multilingual Social Media Data","volume":"16","author":"Almalki","year":"2025","journal-title":"Int. J. Adv. Comput. Sci. Appl."},{"key":"ref_8","first-page":"1","article-title":"Sentiment analysis using XLM-R transformer and zero-shot transfer learning on resource-poor Indian language","volume":"20","author":"Kumar","year":"2021","journal-title":"Trans. Asian Low-Resour. Lang. Inf. Process."},{"key":"ref_9","doi-asserted-by":"crossref","unstructured":"Li, X., and Zhang, K. (2025). Contrastive Learning Pre-Training and Quantum Theory for Cross-Lingual Aspect-Based Sentiment Analysis. Entropy, 27.","DOI":"10.3390\/e27070713"},{"key":"ref_10","doi-asserted-by":"crossref","first-page":"1","DOI":"10.1145\/3649451","article-title":"Financial sentiment analysis: Techniques and applications","volume":"56","author":"Du","year":"2024","journal-title":"ACM Comput. Surv."},{"key":"ref_11","unstructured":"Yang, S., Rosenfeld, J., and Makutonin, J. (2018). Financial aspect-based sentiment analysis using deep representations. arXiv."},{"key":"ref_12","doi-asserted-by":"crossref","first-page":"782","DOI":"10.1002\/asi.23062","article-title":"Good debt or bad debt: Detecting semantic orientations in economic texts","volume":"65","author":"Malo","year":"2014","journal-title":"J. Assoc. Inf. Sci. Technol."},{"key":"ref_13","unstructured":"Yang, Y., Uy, M.C.S., and Huang, A. (2020). Finbert: A pretrained language model for financial communications. arXiv."},{"key":"ref_14","doi-asserted-by":"crossref","first-page":"2244","DOI":"10.1016\/j.procs.2025.04.477","article-title":"Enhancing Financial Sentiment Analysis: Integrating the LoughranMcDonald Dictionary with BERT for Advanced Market Predictive Insights","volume":"258","author":"Sheetal","year":"2025","journal-title":"Procedia Comput. Sci."},{"key":"ref_15","doi-asserted-by":"crossref","first-page":"806","DOI":"10.1111\/1911-3846.12832","article-title":"FinBERT: A large language model for extracting information from financial text","volume":"40","author":"Huang","year":"2023","journal-title":"Contemp. Account. Res."},{"key":"ref_16","unstructured":"Doddapaneni, S., Ramesh, G., Khapra, M.M., Kunchukuttan, A., and Kumar, P. (2021). A primer on pretrained multilingual language models. arXiv."},{"key":"ref_17","doi-asserted-by":"crossref","unstructured":"Kalia, I., Singh, P., and Kumar, A. (2024). Domain Adaptation for NER Using mBERT. Proceedings of the International Conference on Innovations in Computational Intelligence and Computer Vision, Springer.","DOI":"10.1007\/978-981-97-6992-6_14"},{"key":"ref_18","doi-asserted-by":"crossref","unstructured":"Ma, Y. (2024, January 30\u201331). Cross-language Text Generation Using mBERT and XLM-R: English-Chinese Translation Task. Proceedings of the 2024 International Conference on Machine Intelligence and Digital Applications, Ningbo, China.","DOI":"10.1145\/3662739.3672320"},{"key":"ref_19","unstructured":"Rizvi, A., Thamindu, N., Adhikari, A., Senevirathna, W., Kasthurirathna, D., and Abeywardhana, L. (2025). Enhancing Multilingual Sentiment Analysis with Explainability for Sinhala, English, and Code-Mixed Content. arXiv."},{"key":"ref_20","unstructured":"Ri, R., Kiyono, S., and Takase, S. (2024). Self-translate-train: Enhancing cross-lingual transfer of large language models via inherent capability. arXiv."},{"key":"ref_21","doi-asserted-by":"crossref","unstructured":"Liu, J., Yang, Y., and Tam, K.Y. (2024). Beyond surface similarity: Detecting subtle semantic shifts in financial narratives. arXiv.","DOI":"10.18653\/v1\/2024.findings-naacl.168"},{"key":"ref_22","doi-asserted-by":"crossref","first-page":"701","DOI":"10.1007\/s10489-025-06578-z","article-title":"AdaFT: An efficient domain-adaptive fine-tuning framework for sentiment analysis in chinese financial texts","volume":"55","author":"Yan","year":"2025","journal-title":"Appl. Intell."},{"key":"ref_23","first-page":"6827","article-title":"What makes for good views for contrastive learning?","volume":"33","author":"Tian","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_24","unstructured":"Zhang, H., and Cao, Y. (2024). Understanding the Benefits of SimCLR Pre-Training in Two-Layer Convolutional Neural Networks. arXiv."},{"key":"ref_25","doi-asserted-by":"crossref","unstructured":"He, K., Fan, H., Wu, Y., Xie, S., and Girshick, R. (2020, January 13\u201319). Momentum contrast for unsupervised visual representation learning. Proceedings of the IEEE\/CVF Conference on Computer Vision and Pattern Recognition, Seattle, WA, USA.","DOI":"10.1109\/CVPR42600.2020.00975"},{"key":"ref_26","doi-asserted-by":"crossref","unstructured":"Gao, T., Yao, X., and Chen, D. (2021). Simcse: Simple contrastive learning of sentence embeddings. arXiv.","DOI":"10.18653\/v1\/2021.emnlp-main.552"},{"key":"ref_27","doi-asserted-by":"crossref","first-page":"20","DOI":"10.1038\/s43586-021-00018-1","article-title":"CLIP and complementary methods","volume":"1","author":"Hafner","year":"2021","journal-title":"Nat. Rev. Methods Prim."},{"key":"ref_28","unstructured":"Feng, F., Yang, Y., Cer, D., Arivazhagan, N., and Wang, W. (2020). Language-agnostic BERT sentence embedding. arXiv."},{"key":"ref_29","doi-asserted-by":"crossref","unstructured":"Wang, Y.S., Wu, A., and Neubig, G. (2022). English contrastive learning can learn universal cross-lingual sentence embeddings. arXiv.","DOI":"10.18653\/v1\/2022.emnlp-main.621"},{"key":"ref_30","doi-asserted-by":"crossref","first-page":"77739","DOI":"10.1109\/ACCESS.2024.3406993","article-title":"Improving BERTScore for machine translation evaluation through contrastive learning","volume":"12","author":"Tang","year":"2024","journal-title":"IEEE Access"},{"key":"ref_31","doi-asserted-by":"crossref","unstructured":"Pires, T., Schlinger, E., and Garrette, D. (2019). How multilingual is multilingual BERT?. arXiv.","DOI":"10.18653\/v1\/P19-1493"},{"key":"ref_32","doi-asserted-by":"crossref","unstructured":"Conneau, A., Khandelwal, K., Goyal, N., Chaudhary, V., Wenzek, G., Guzm\u00e1n, F., Grave, E., Ott, M., Zettlemoyer, L., and Stoyanov, V. (2019). Unsupervised cross-lingual representation learning at scale. arXiv.","DOI":"10.18653\/v1\/2020.acl-main.747"},{"key":"ref_33","doi-asserted-by":"crossref","unstructured":"Yu, S., Sun, Q., Zhang, H., and Jiang, J. (2022). Translate-Train Embracing Translationese Artifacts, Association for Computational Linguistics.","DOI":"10.18653\/v1\/2022.acl-short.40"}],"container-title":["Computers"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/2073-431X\/14\/8\/338\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T18:30:27Z","timestamp":1760034627000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/2073-431X\/14\/8\/338"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8,19]]},"references-count":33,"journal-issue":{"issue":"8","published-online":{"date-parts":[[2025,8]]}},"alternative-id":["computers14080338"],"URL":"https:\/\/doi.org\/10.3390\/computers14080338","relation":{},"ISSN":["2073-431X"],"issn-type":[{"value":"2073-431X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8,19]]}}}