{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,25]],"date-time":"2026-03-25T15:52:33Z","timestamp":1774453953361,"version":"3.50.1"},"reference-count":33,"publisher":"Association for Computing Machinery (ACM)","issue":"4","license":[{"start":{"date-parts":[[2025,3,23]],"date-time":"2025-03-23T00:00:00Z","timestamp":1742688000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/www.acm.org\/publications\/policies\/copyright_policy#Background"}],"funder":[{"name":"Shanghai Technical Service Center of Science and Engineering Computing, Shanghai University"}],"content-domain":{"domain":["dl.acm.org"],"crossmark-restriction":true},"short-container-title":["ACM Trans. Asian Low-Resour. Lang. Inf. Process."],"published-print":{"date-parts":[[2025,4,30]]},"abstract":"<jats:p>Chinese poetry, renowned for its elegance and simplicity, is a hallmark of Chinese culture. While neural networks have made significant advancements in generating poetry, balancing diversity with adherence to rigid structural formats remains a challenge. Research indicates that factors such as themes, emotions (e.g., happiness, sadness), and sentiments (e.g., positive, negative) play a crucial role in poetic creation, influencing both the diversity and quality of the generated content. In this paper, we propose MixSong, an autoregressive language model based on the Transformer architecture, designed to incorporate a wide range of conditional factors. MixSong utilizes adversarial training to integrate these factors, enabling the model to implicitly learn distributional information in the latent space. Additionally, we introduce several uniquely customized symbol sets, including paragraph identifiers, position identifiers, rhyme identifiers, tune identifiers, and conditional distinctive identifiers. These symbols help MixSong effectively capture and enforce the constraints necessary for generating high-quality poetry. Extensive experimental results demonstrate that MixSong significantly outperforms existing models in both automatic metrics and human evaluations, achieving notable improvements in both diversity and quality of the generated poetry.<\/jats:p>","DOI":"10.1145\/3718331","type":"journal-article","created":{"date-parts":[[2025,2,17]],"date-time":"2025-02-17T11:32:36Z","timestamp":1739791956000},"page":"1-16","update-policy":"https:\/\/doi.org\/10.1145\/crossmark-policy","source":"Crossref","is-referenced-by-count":2,"title":["MixSong: Diverse and Strictly Formatted Chinese Poetry Generation"],"prefix":"10.1145","volume":"24","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-1216-279X","authenticated-orcid":false,"given":"Xinglong","family":"Song","sequence":"first","affiliation":[{"name":"Shanghai Film Academy, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-3501-7027","authenticated-orcid":false,"given":"Changlin","family":"Song","sequence":"additional","affiliation":[{"name":"Shanghai Film Academy, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-2434-4334","authenticated-orcid":false,"given":"Haolu","family":"Yu","sequence":"additional","affiliation":[{"name":"Shanghai Film Academy, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6419-9018","authenticated-orcid":false,"given":"Yonghua","family":"Zhu","sequence":"additional","affiliation":[{"name":"Shanghai Film Academy, Shanghai University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7967-8962","authenticated-orcid":false,"given":"Hong","family":"Yao","sequence":"additional","affiliation":[{"name":"School of Information Technology, Shanghai Jianqiao University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"320","published-online":{"date-parts":[[2025,3,23]]},"reference":[{"key":"e_1_3_2_2_2","unstructured":"J. Devlin M. W. Chang K. Lee et al. 2018. BERT: Pre-training of deep bidirectional transformers for language understanding. arxiv preprint arxiv:1810.04805 (2018)."},{"key":"e_1_3_2_3_2","doi-asserted-by":"crossref","unstructured":"A. Fan M. Lewis and Y. Dauphin. 2018. Hierarchical neural story generation. arxiv preprint arxiv:1805.04833 (2018).","DOI":"10.18653\/v1\/P18-1082"},{"key":"e_1_3_2_4_2","doi-asserted-by":"publisher","DOI":"10.1016\/S0950-7051(01)00095-8"},{"key":"e_1_3_2_5_2","doi-asserted-by":"crossref","unstructured":"H. Gon\u00e7alo Oliveira and A. Cardoso. 2014. Poetry generation with PoeTryMe. Computational Creativity Research: Towards Creative Machines. Paris: Atlantis Press 2014: 243\u2013266.","DOI":"10.2991\/978-94-6239-085-0_12"},{"key":"e_1_3_2_6_2","first-page":"27","article-title":"Generative adversarial nets","author":"Goodfellow I.","year":"2014","unstructured":"I. Goodfellow, J. Pouget-Abadie, M. Mirza et al. 2014. Generative adversarial nets. Advances in Neural Information Processing Systems, (2014), 27.","journal-title":"Advances in Neural Information Processing Systems"},{"key":"e_1_3_2_7_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v26i1.8344"},{"key":"e_1_3_2_8_2","unstructured":"J. Li M. Galley C. Brockett et al. 2015. A diversity-promoting objective function for neural conversation models. arxiv preprint arxiv:1510.03055 (2015)."},{"key":"e_1_3_2_9_2","unstructured":"I. Loshchilov and F. Hutter. 2016. SGDR: Stochastic gradient descent with warm restarts. arxiv preprint arxiv:1608.03983 (2016)."},{"key":"e_1_3_2_10_2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1423"},{"key":"e_1_3_2_11_2","unstructured":"Y. Liao Y. Wang Q. Liu et al. 2019. GPT-based generation for classical Chinese poetry. arXiv preprint arXiv:1907.00151 (2019)."},{"key":"e_1_3_2_12_2","unstructured":"P. Li H. Zhang X. Liu et al. 2004. SongNet: Rigid formats controlled text generation. arxiv preprint arxiv:2004.08022 2020."},{"key":"e_1_3_2_13_2","unstructured":"H. M. Manurung. 2004. An evolutionary algorithm approach to poetry generation. Ph.D. Dissertation. School of Informatics University of Edinburgh."},{"key":"e_1_3_2_14_2","unstructured":"T. Mikolov K. Chen G. Corrado et al. 2013. Efficient estimation of word representations in vector space. arxiv preprint arxiv:1301.3781 (2013)."},{"key":"e_1_3_2_15_2","unstructured":"S. Mohamed and B. Lakshminarayanan. 2016. Learning in implicit generative models. arxiv preprint arxiv:1610.03483 (2016)."},{"key":"e_1_3_2_16_2","doi-asserted-by":"publisher","DOI":"10.1073\/pnas.0706851105"},{"key":"e_1_3_2_17_2","unstructured":"M. Rosca B. Lakshminarayanan D. Warde-Farley et al. 2017. Variational approaches for auto-encoding generative adversarial networks. arxiv preprint arxiv:1706.04987 (2017)."},{"key":"e_1_3_2_18_2","unstructured":"A. Radford K. Narasimhan T. Salimans et al. 2018. Improving Language Understanding by Generative Pre-Training. Technical Report. OpenAI. https:\/\/cdn.openai.com\/research-covers\/language-unsupervised\/language_understanding_paper.pdf"},{"issue":"8","key":"e_1_3_2_19_2","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford A.","year":"2019","unstructured":"A. Radford, J. Wu, R. Child et al. 2019. Language models are unsupervised multitask learners. OpenAI Blog 1, 8 (2019), 9.","journal-title":"OpenAI Blog"},{"key":"e_1_3_2_20_2","doi-asserted-by":"publisher","DOI":"10.1007\/BF00992696"},{"key":"e_1_3_2_21_2","unstructured":"Z. Wang W. He H. Wu et al. 2016. Chinese poetry generation with planning based neural network. arxiv preprint arxiv:1610.09889 (2016)."},{"key":"e_1_3_2_22_2","doi-asserted-by":"crossref","first-page":"34","DOI":"10.1007\/978-3-319-49685-6_4","article-title":"Can machine generate traditional Chinese poetry? A Feigenbaum test","author":"Wang Q.","year":"2016","unstructured":"Q. Wang, T. Luo, and D. Wang. 2016. Can machine generate traditional Chinese poetry? A Feigenbaum test. Advances in Brain Inspired Cognitive Systems: 8th International Conference, BICS 2016, Beijing, China, November 28-30, 2016, Proceedings 8. Springer International Publishing, 2016: 34\u201346.","journal-title":"Advances in Brain Inspired Cognitive Systems: 8th International Conference, BICS 2016, Beijing, China, November 28-30, 2016, Proceedings 8. Springer International Publishing"},{"key":"e_1_3_2_23_2","unstructured":"Q. Wang T. Luo D. Wang et al. 2016. Chinese song iambics generation with neural attention-based model. arxiv preprint arxiv:1604.06274 (2016)."},{"key":"e_1_3_2_24_2","unstructured":"Xiaoyuan Yi Ruoyu Li and Maosong Sun. 2016. Generating Chinese classical poems with RNN encoder-decoder. arXiv preprint arXiv:1604.01537 (2016)."},{"key":"e_1_3_2_25_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v31i1.10804"},{"key":"e_1_3_2_26_2","unstructured":"X. Yang X. Lin S. Suo et al. 2017. Generating thematic Chinese poetry using conditional variational autoencoders with hybrid decoders. arxiv preprint arxiv:1711.07632 (2017)."},{"key":"e_1_3_2_27_2","unstructured":"X. Yi R. Li and M. Sun. 2018. Chinese poetry generation with a salient-clue mechanism. arxiv preprint arxiv:1809.04313 (2018)."},{"key":"e_1_3_2_28_2","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D18-1353"},{"key":"e_1_3_2_29_2","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6488"},{"key":"e_1_3_2_30_2","doi-asserted-by":"publisher","DOI":"10.1145\/3593805"},{"key":"e_1_3_2_31_2","doi-asserted-by":"publisher","DOI":"10.3115\/v1\/D14-1074"},{"key":"e_1_3_2_32_2","doi-asserted-by":"crossref","unstructured":"T. Zhao R. Zhao and M. Eskenazi. 2017. Learning discourse-level diversity for neural dialog models using conditional variational autoencoders. arxiv preprint arxiv:1703.10960 (2017).","DOI":"10.18653\/v1\/P17-1061"},{"key":"e_1_3_2_33_2","doi-asserted-by":"crossref","unstructured":"B. Zhang D. Xiong J. Su et al. 2016. Variational neural machine translation. arXiv preprint arXiv:1605.07869 2016.","DOI":"10.18653\/v1\/D16-1050"},{"key":"e_1_3_2_34_2","article-title":"Generative adversarial and self-supervised dehazing network","author":"Zhang S.","year":"2023","unstructured":"S. Zhang, X. Zhang, S. Wan et al. 2023. Generative adversarial and self-supervised dehazing network. IEEE Transactions on Industrial Informatics, (2023).","journal-title":"IEEE Transactions on Industrial Informatics"}],"container-title":["ACM Transactions on Asian and Low-Resource Language Information Processing"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3718331","content-type":"unspecified","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/dl.acm.org\/doi\/pdf\/10.1145\/3718331","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,19]],"date-time":"2025-06-19T01:57:34Z","timestamp":1750298254000},"score":1,"resource":{"primary":{"URL":"https:\/\/dl.acm.org\/doi\/10.1145\/3718331"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3,23]]},"references-count":33,"journal-issue":{"issue":"4","published-print":{"date-parts":[[2025,4,30]]}},"alternative-id":["10.1145\/3718331"],"URL":"https:\/\/doi.org\/10.1145\/3718331","relation":{},"ISSN":["2375-4699","2375-4702"],"issn-type":[{"value":"2375-4699","type":"print"},{"value":"2375-4702","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,3,23]]},"assertion":[{"value":"2024-06-18","order":0,"name":"received","label":"Received","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-02-08","order":2,"name":"accepted","label":"Accepted","group":{"name":"publication_history","label":"Publication History"}},{"value":"2025-03-23","order":3,"name":"published","label":"Published","group":{"name":"publication_history","label":"Publication History"}}]}}