pith. sign in

arxiv: 2605.22654 · v1 · pith:2KR2BRGEnew · submitted 2026-05-21 · 💻 cs.CL · cs.CV

Seeing the Poem: Image-Semantic Detection of AI-Generated Modern Chinese Poetry with MLLMs

Shanshan Wang , Fengying Ye , Hanjia Lyu , Caiwen Gou , Junchao Wu , Jingming Yao , Chengzhong Xu , Jiebo Luo
show 1 more author
Derek F. Wong
This is my paper

Pith reviewed 2026-05-22 05:38 UTC · model grok-4.3

classification 💻 cs.CL cs.CV
keywords AI generated poetrymodern Chinese poetrymultimodal detectionLLM detectorsimage semantic analysispoetry authenticity
0
0 comments X

The pith

Adding images that reflect poem content improves LLM detection of AI-generated modern Chinese poetry

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper proposes using images that capture the meaning, imagery, and feeling of modern Chinese poems to help large language models detect whether the poetry was generated by AI. Earlier studies showed LLMs struggle as detectors for AI text in general, but this work focuses specifically on modern Chinese poetry and finds that combining text with semantic images creates a stronger detection signal. Experiments show this image-semantic approach lifts performance above plain-text LLM detectors and even beats the strong traditional baseline RoBERTa, with the Gemini model reaching 85.65 percent Macro-F1. A reader might care because reliable detection supports the integrity of literary traditions as generative AI spreads into poetry writing.

Core claim

By incorporating images that reflect the content of the poetry, the image-semantic guided method allows LLMs to integrate complementary information on meaning, imagery, and feeling, leading to more accurate detection of AI-generated modern Chinese poetry compared to text-only methods.

What carries the argument

Image-semantic guided poetry detection method that forms complementary judgments from poem text and matching images

If this is right

  • LLM detectors outperform text-only baselines on multiple AI-generated poetry datasets
  • The Gemini-based detector reaches state-of-the-art Macro-F1 of 85.65%
  • The method surpasses the best traditional detector RoBERTa
  • Performance gains are observed across different LLMs

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

  • This technique could extend to detecting AI content in other image-rich creative fields such as visual art descriptions or song lyrics.
  • Poetry detection might benefit from always generating an illustrative image as a first step before analysis.
  • Future detectors may need to account for how image generation models themselves introduce patterns that could be exploited or masked.

Load-bearing premise

Images can be generated or selected to match the poetry's meaning, imagery, and feeling closely enough to aid detection without introducing their own biases or artifacts.

What would settle it

Running the image-semantic detector and a plain-text detector on a new collection of human and AI-written modern Chinese poems and finding no accuracy advantage for the image version.

Figures

Figures reproduced from arXiv: 2605.22654 by Caiwen Gou, Chengzhong Xu, Derek F. Wong, Fengying Ye, Hanjia Lyu, Jiebo Luo, Jingming Yao, Junchao Wu, Shanshan Wang.

Figure 1
Figure 1. Figure 1: The relationship between human poetry and [PITH_FULL_IMAGE:figures/full_fig_p001_1.png] view at source ↗
Figure 2
Figure 2. Figure 2: Comparison between the framework of the traditional detection method and our proposed IMAGINE [PITH_FULL_IMAGE:figures/full_fig_p002_2.png] view at source ↗
read the original abstract

Previous detection studies have shown that LLMs cannot be effectively used as detectors, but these studies have not addressed modern Chinese poetry. Moreover, no relevant research has explored the performance of LLMs in detecting modern Chinese poetry. This paper evaluates and enhances the performance of LLMs as detectors for modern Chinese poetry, and proposes an image-semantic guided poetry detection method. Compared with traditional detection approaches, our method innovatively incorporates images that reflect the content of the poetry. Through example-driven approaches, our method effectively integrates information such as meaning, imagery, and feeling from the image, then forms a complementary judgment with the poem text. Experimental results demonstrate that the LLM detectors based on our method outperform baseline detectors based on plain text, and even surpass the best-performing traditional detector, RoBERTa. The Gemini detector using our method achieves a Macro-F1 score of 85.65%, reaching the state-of-the-art level. The performance improvements of different LLM detectors on multiple LLMs-generated data prove the effectiveness of our method.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Referee Report

2 major / 2 minor

Summary. The paper proposes an image-semantic guided detection method for AI-generated modern Chinese poetry that augments LLM/MLLM prompts with images reflecting poem content (meaning, imagery, feeling). It reports that this multimodal approach yields higher detection performance than text-only LLM baselines and surpasses the RoBERTa baseline, with the Gemini-based detector reaching 85.65% Macro-F1.

Significance. If the performance gain is shown to arise from genuine semantic complementarity rather than low-level image artifacts, the work would provide a concrete multimodal technique for detecting generated creative text and could inform future detectors that exploit imagery in poetry and similar domains.

major comments (2)
  1. [Method / Experimental Setup] The central experimental claim (abstract and results) that images supply complementary semantic information rests on an unspecified image generation or selection process. No details are given on the text-to-image model, prompt construction, or any control condition that would isolate semantic content from correlated artifacts (texture inconsistencies, prompt leakage). This directly affects interpretability of the 85.65% Macro-F1 and the outperformance over RoBERTa.
  2. [Experiments] Dataset construction is described only at a high level; the manuscript does not report how human-written vs. LLM-generated poems were collected, balanced, or split, nor any statistical significance tests or error analysis that would substantiate the cross-model performance gains.
minor comments (2)
  1. [Method] Notation for the image-semantic fusion step could be clarified with a short pseudocode or diagram to show exactly how image features are combined with text in the MLLM prompt.
  2. [Abstract / Method] The abstract states that the method 'forms a complementary judgment'; a concrete example of a prompt template and the resulting MLLM output would help readers reproduce the integration step.

Simulated Author's Rebuttal

2 responses · 0 unresolved

We thank the referee for the constructive feedback, which has helped us identify areas where the manuscript requires greater clarity and detail. We address each major comment below and have prepared revisions to strengthen the experimental description and reproducibility.

read point-by-point responses

  1. Referee: [Method / Experimental Setup] The central experimental claim (abstract and results) that images supply complementary semantic information rests on an unspecified image generation or selection process. No details are given on the text-to-image model, prompt construction, or any control condition that would isolate semantic content from correlated artifacts (texture inconsistencies, prompt leakage). This directly affects interpretability of the 85.65% Macro-F1 and the outperformance over RoBERTa.

    Authors: We agree that the original manuscript described the image-augmentation process at too high a level, limiting readers' ability to assess whether gains derive from semantic content or from low-level artifacts. In the revised version we have added a dedicated subsection that specifies the text-to-image model, the exact prompt templates used to translate poem semantics into images, and a control experiment that compares performance with semantically faithful images against images generated from shuffled or artifact-only prompts. These additions directly address the interpretability concern raised. revision: yes

  2. Referee: [Experiments] Dataset construction is described only at a high level; the manuscript does not report how human-written vs. LLM-generated poems were collected, balanced, or split, nor any statistical significance tests or error analysis that would substantiate the cross-model performance gains.

    Authors: We acknowledge that the dataset section was insufficiently detailed. The revised manuscript now contains an expanded data section that reports the sources of human-written poems, the specific LLMs and generation settings used to create the AI poems, the balancing and splitting procedures, and the final dataset sizes. We have also added statistical significance testing (paired t-tests and McNemar's test) for all reported improvements and a concise error-analysis subsection that categorizes the remaining misclassifications. revision: yes

Circularity Check

0 steps flagged

Empirical evaluation with external baselines is self-contained

full rationale

The paper describes an empirical method that augments poem text with images reflecting its content and evaluates LLM-based detectors (including Gemini) against plain-text baselines and the external RoBERTa model, reporting a Macro-F1 of 85.65%. No equations, fitted parameters renamed as predictions, self-definitional constructs, or load-bearing self-citations appear in the abstract or method framing. The performance claims rest on direct experimental comparisons to independent detectors rather than reducing to the method's own inputs by construction.

Axiom & Free-Parameter Ledger

1 free parameters · 1 axioms · 0 invented entities

The central claim rests on the domain assumption that visual semantic representations can complement textual analysis for detection; limited free parameters are evident from the abstract description.

free parameters (1)
  • Image generation or selection process
    The specific prompts or method used to create images reflecting poetry content is a design choice that affects the complementary judgment.
axioms (1)
  • domain assumption Images can capture and convey meaning, imagery, and feeling from poetry text in a way useful for AI detection
    Invoked when stating that the method integrates information from the image to form a complementary judgment with the poem text.

pith-pipeline@v0.9.0 · 5740 in / 1298 out tokens · 43503 ms · 2026-05-22T05:38:16.728683+00:00 · methodology

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

What do these tags mean?
matches
The paper's claim is directly supported by a theorem in the formal canon.
supports
The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends
The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses
The paper appears to rely on the theorem as machinery.
contradicts
The paper's claim conflicts with a theorem or certificate in the canon.
unclear
Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

106 extracted references · 106 canonical work pages · 12 internal anchors

  1. [3]

    James Betker, Gabriel Goh, Li Jing, Tim Brooks, Jianfeng Wang, Linjie Li, Long Ouyang, Juntang Zhuang, Joyce Lee, Yufei Guo, and 1 others. 2023. https://cdn.openai.com/papers/dall-e-3.pdf Improving image generation with better captions . Computer Science., 2(3):8

    work page 2023

  2. [4]

    Stefan Blohm, Valentin Wagner, Matthias Schlesewsky, and Winfried Menninghaus. 2018. https://www.sciencedirect.com/science/article/abs/pii/S0304422X17300864?casa_token=fZLWD530VLEAAAAA:6hL3soz_GX_rtn_KIC2gg3YIGOcagdgzEMJFMWje77zdeWHbsvoRzsf8I0UlPIWWAYQS_lqgtLaY Sentence judgments and the grammar of poetry: Linking linguistic structure and poetic effect . ...

    work page 2018

  3. [6]

    Tianqi Chen. 2016. Xgboost: A scalable tree boosting system. Cornell University

    work page 2016

  4. [8]

    twist the neck of grammar

    Zhongyi Chen. 2012. https://xueshu.baidu.com/usercenter/paper/show?paperid=c974e7f2cdc678422498efa431cdf15e&site=xueshu_se Why "twist the neck of grammar" - a study of modern poetic rhetoric . Journal of Central China Normal University: Humanities and Social Sciences Edition, 51(1):7

    work page 2012

  5. [10]

    Kohinoor Darda, Marion Carre, and Emily Cross. 2023. https://royalsocietypublishing.org/doi/10.1098/rsos.220915 Value attributed to text-based archives generated by artificial intelligence . Royal Society Open Science, 10(2):220915

    work page doi:10.1098/rsos.220915 2023

  6. [11]

    Cheng Deng. 2007. https://xueshu.baidu.com/usercenter/paper/show?paperid=9e2401da68a441c1d5e3c691c144a586&site=xueshu_se Dilemma and solution: Reflections on current new poetry . Literary Review, (3):4

    work page 2007

  7. [12]

    Sebastian Gehrmann, Hendrik Strobelt, and Alexander Rush. 2019. https://aclanthology.org/P19-3019/ GLTR : Statistical detection and visualization of generated text . In Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics: System Demonstrations, pages 111--116. Association for Computational Linguistics

    work page 2019

  8. [15]

    Kadhim Hayawi, Sakib Shahriar, and Sujith Samuel Mathew. 2024. https://arxiv.org/abs/2307.12166 The imitation game: Detecting human and ai-generated texts in the era of chatgpt and bard . Journal of Information Science, page 01655515241227531

    work page arXiv 2024

  9. [16]

    Min Hu. 2003. https://xueshu.baidu.com/ndscholar/browse/detail?paperid=8777ea800ec979a294010a8b20201a30&site=xueshu_se On the imagery beauty of scene-emotion blending in classical chinese poetry . Jiangxi Social Sciences, (4):3

    work page 2003

  10. [17]

    quasi-text

    Junming Huo. 2020. https://xueshu.baidu.com/usercenter/paper/show?paperid=1q0y0rk03s1v0e00vc2v0840f8747493&site=xueshu_se Clone li bai and 100 trillion poems - the "quasi-text" production and possible prospects of ai poetry . Southern Literature, (4):5

    work page 2020

  11. [19]

    Maurice Jakesch, Jeffrey T Hancock, and Mor Naaman. 2023. https://www.pnas.org/doi/10.1073/pnas.2208839120 Human heuristics for ai-generated language are flawed . Proceedings of the National Academy of Sciences, 120(11):e2208839120

    work page doi:10.1073/pnas.2208839120 2023

  12. [20]

    H Tin Kam. 1995. Random decision forests. proceedings of 3rd international conference on document analysis and recognition

    work page 1995

  13. [21]

    Nils K \"o bis and Luca D Mossink. 2021. https://arxiv.org/abs/2005.09980 Artificial intelligence versus maya angelou: Experimental evidence that people cannot differentiate ai-generated from human-written poetry . Computers in human behavior, 114:106553

    work page arXiv 2021

  14. [22]

    Zhixin Lai, Xuesheng Zhang, and Suiyao Chen. 2024. https://ieeexplore.ieee.org/abstract/document/10651296 Adaptive ensembles of fine-tuned transformers for llm-generated text detection . In 2024 International Joint Conference on Neural Networks (IJCNN), pages 1--7. IEEE

    work page arXiv 2024

  15. [23]

    Tian Lan, Jiang Li, Yemin Wang, Xu Liu, Xiangdong Su, and Guanglai Gao. 2025 a . https://aclanthology.org/2025.emnlp-main.105/ F B ench: An open-ended fairness evaluation benchmark for LLM s with factuality considerations . In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 2031--2046. Association for Computat...

    work page 2025

  16. [24]

    Tian Lan, Xiangdong Su, Xu Liu, Ruirui Wang, Ke Chang, Jiang Li, and Guanglai Gao. 2025 b . https://aclanthology.org/2025.findings-acl.313/ M c BE : A multi-task C hinese bias evaluation benchmark for large language models . In Findings of the Association for Computational Linguistics: ACL 2025, pages 6033--6056

    work page 2025

  17. [25]

    Haotian Li, Jiatao Zhu, Sichen Cao, Xiangyu Li, Jiajun Zeng, and Peng Wang. 2021. https://link.springer.com/chapter/10.1007/978-3-030-73197-7_43 Poetic expression through scenery: sentimental chinese classical poetry generation from images . In International Conference on Database Systems for Advanced Applications, pages 629--637. Springer

    work page doi:10.1007/978-3-030-73197-7_43 2021

  18. [27]

    Dianqing Lin, Aruukhan, Hongxu Hou, Shuo Sun, Wei Chen, Yichen Yang, and Guodong Shi. 2025. https://aclanthology.org/2025.emnlp-main.1179/ Can large language models translate unseen languages in underrepresented scripts? In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 23137--23150. Association for Computati...

    work page 2025

  19. [29]

    Bei Liu, Jianlong Fu, Makoto P Kato, and Masatoshi Yoshikawa. 2018. https://dl.acm.org/doi/abs/10.1145/3240508.3240587 Beyond narrative description: Generating poetry from images by multi-adversarial training . In Proceedings of the 26th ACM international conference on Multimedia, pages 783--791

    work page doi:10.1145/3240508.3240587 2018

  20. [31]

    Hanjia Lyu, Jiebo Luo, Jian Kang, and Allison Koenecke. 2025. https://doi.org/10.1145/3715275.3732182 Characterizing bias: Benchmarking large language models in simplified versus traditional chinese . In Proceedings of the 2025 ACM Conference on Fairness, Accountability, and Transparency, page 2815–2846. Association for Computing Machinery

    work page doi:10.1145/3715275.3732182 2025

  21. [33]

    Trung T Nguyen, Amartya Hatua, and Andrew H Sung. 2023. https://ieeexplore.ieee.org/abstract/document/10316132 How to detect ai-generated texts? In 2023 IEEE 14th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON), pages 0464--0471. IEEE

    work page arXiv 2023

  22. [34]

    OpenAI. 2025. https://platform.openai.com/docs/models/gpt-image-1 Gpt-image-1 model documentation

    work page 2025

  23. [35]

    Brian Porter and Edouard Machery. 2024. https://www.nature.com/articles/s41598-024-76900-1 Ai-generated poetry is indistinguishable from human-written poetry and is rated more favorably . Scientific Reports, 14(1):26133

    work page 2024

  24. [36]

    Xuelian Ren, Xiaolong Chai, and Mingzhi Mao. 2023. https://dl.acm.org/doi/abs/10.1145/3656766.3656790 Generating chinese poetry from images based on deep learning . In Proceedings of the 2023 3rd International Conference on Big Data, Artificial Intelligence and Risk Management, pages 134--138

    work page doi:10.1145/3656766.3656790 2023

  25. [40]

    Wong, Runzhe Zhan, Lidia S

    Wai Lei Song, Haoyun Xu, Derek F. Wong, Runzhe Zhan, Lidia S. Chao, and Shanshan Wang. 2023. https://aclanthology.org/2023.mtsummit-research.27/ Towards zero-shot multilingual poetry translation . In Proceedings of Machine Translation Summit XIX, Vol. 1: Research Track, pages 324--335. Asia-Pacific Association for Machine Translation

    work page 2023

  26. [42]

    Deming Wang. 2003. https://xueshu.baidu.com/ndscholar/browse/search?wd=中国古代早期诗歌创作中借景抒情手法形成的三个阶段&paperid=e6e1cb61ab39614e01a1e14a9bfbe90e&site=xueshu_se The three stages in the formation of the technique of expressing emotions through describing scenes in early ancient chinese poetry creation . Journal of Baise University, (5):42--46

    work page 2003

  27. [43]

    Shanshan Wang, Derek Wong, Jingming Yao, and Lidia Chao. 2024. https://aclanthology.org/2024.acl-long.756/ What is the best way for chatgpt to translate poetry? In Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 14025--14043

    work page 2024

  28. [44]

    Wong, Jingming Yao, and Lidia S

    Shanshan Wang, Derek F. Wong, Jingming Yao, and Lidia S. Chao. 2026. https://aclanthology.org/2026.findings-eacl.216/ Can C hat GPT really understand M odern C hinese poetry? In Findings of the A ssociation for C omputational L inguistics: EACL 2026 , pages 4152--4162. Association for Computational Linguistics

    work page 2026

  29. [45]

    Wong, Jingming Yao, and Lidia S

    Shanshan Wang, Junchao Wu, Fengying Ye, Derek F. Wong, Jingming Yao, and Lidia S. Chao. 2025. https://aclanthology.org/2025.findings-emnlp.507/ Benchmarking the detection of LLM s-generated M odern C hinese poetry . In Findings of the Association for Computational Linguistics: EMNLP 2025, pages 9533--9552. Association for Computational Linguistics

    work page 2025

  30. [47]

    Junchao Wu, Shu Yang, Runzhe Zhan, Yulin Yuan, Lidia Sam Chao, and Derek Fai Wong. 2025 b . https://aclanthology.org/2025.cl-1.8/ A survey on llm-generated text detection: Necessity, methods, and future directions . Computational Linguistics, pages 1--66

    work page 2025

  31. [49]

    Zhengrong Xiong. 2001. https://xueshu.baidu.com/ndscholar/browse/detail?paperid=4c41d54cda2477f3b9a97c0b5bd6922c&site=xueshu_se Why so many spring sorrows and autumn regrets: A preliminary exploration of the creative mentality of ancient chinese poets . Journal of Huainan Normal University, (4):2

    work page 2001

  32. [50]

    Linli Xu, Liang Jiang, Chuan Qin, Zhe Wang, and Dongfang Du. 2018. https://ojs.aaai.org/index.php/AAAI/article/view/12001 How images inspire poems: Generating classical chinese poetry from images with memory networks . In Proceedings of the AAAI Conference on Artificial Intelligence, volume 32

    work page 2018

  33. [52]

    Xiaoyuan Yi, Maosong Sun, Ruoyu Li, and Zonghan Yang. 2018. https://www.ijcai.org/proceedings/2018/0633.pdf Chinese poetry generation with a working memory model

    work page 2018

  34. [53]

    Daniel Zhang, Bo Ni, Qiyu Zhi, Thomas Plummer, Qi Li, Hao Zheng, Qingkai Zeng, Yang Zhang, and Dong Wang. 2019. https://dl.acm.org/doi/abs/10.1145/3341161.3342885 Through the eyes of a poet: Classical poetry recommendation with visual input on social media . In Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analys...

    work page doi:10.1145/3341161.3342885 2019

  35. [54]

    Jiaqi Zhao, Ting Bai, Yuting Wei, and Bin Wu. 2022. https://link.springer.com/chapter/10.1007/978-981-19-8991-9_26 Poetrybert: Pre-training with sememe knowledge for classical chinese poetry . In International conference on data mining and big data, pages 369--384. Springer

    work page doi:10.1007/978-981-19-8991-9_26 2022

  36. [55]

    ZHIPU. 2024. https://zhipuai.cn/en/devday Zhipu ai devday glm-4 . Accessed: 2024-05-01

    work page 2024

  37. [56]

    2024 International Joint Conference on Neural Networks (IJCNN) , pages=

    Adaptive ensembles of fine-tuned transformers for llm-generated text detection , author=. 2024 International Joint Conference on Neural Networks (IJCNN) , pages=. 2024 , Url=

    work page 2024

  38. [57]

    Proceedings of the National Academy of Sciences , volume=

    Human heuristics for AI-generated language are flawed , author=. Proceedings of the National Academy of Sciences , volume=. 2023 , url=

    work page 2023

  39. [58]

    All That ' s `Human' Is Not Gold: Evaluating Human Evaluation of Generated Text

    Clark, Elizabeth and August, Tal and Serrano, Sofia and Haduong, Nikita and Gururangan, Suchin and Smith, Noah A. All That ' s `Human' Is Not Gold: Evaluating Human Evaluation of Generated Text. Proceedings of the 59th Annual Meeting of the Association for Computational Linguistics and the 11th International Joint Conference on Natural Language Processing...

    work page 2021

  40. [59]

    Scientific Reports , volume=

    AI-generated poetry is indistinguishable from human-written poetry and is rated more favorably , author=. Scientific Reports , volume=. 2024 , url=

    work page 2024

  41. [60]

    Journal of Information Science , pages=

    The imitation game: Detecting human and AI-generated texts in the era of ChatGPT and BARD , author=. Journal of Information Science , pages=. 2024 , url=

    work page 2024

  42. [61]

    arXiv preprint arXiv:2501.03212 , url=

    Leveraging Explainable AI for LLM Text Attribution: Differentiating Human-Written and Multiple LLMs-Generated Text , author=. arXiv preprint arXiv:2501.03212 , url=

    work page arXiv

  43. [62]

    Computational Linguistics , pages=

    A survey on LLM-generated text detection: Necessity, methods, and future directions , author=. Computational Linguistics , pages=. 2025 , url=

    work page 2025

  44. [63]

    克隆体李白"与百万亿首诗——AI诗歌的

    "克隆体李白"与百万亿首诗——AI诗歌的"类文本"生产与可能前景 , author=. 南方文坛 , number=

    work page

  45. [64]

    quasi-text

    Clone Li Bai and 100 trillion poems - the "quasi-text" production and possible prospects of AI poetry , author=. Southern Literature , number=

    work page

  46. [65]

    Chinese poetry generation with a working memory model , author=

    work page

  47. [66]

    International conference on data mining and big data , pages=

    PoetryBERT: Pre-training with sememe knowledge for classical Chinese poetry , author=. International conference on data mining and big data , pages=. 2022 , url=

    work page 2022

  48. [67]

    Computers in human behavior , volume=

    Artificial intelligence versus Maya Angelou: Experimental evidence that people cannot differentiate AI-generated from human-written poetry , author=. Computers in human behavior , volume=. 2021 , url=

    work page 2021

  49. [68]

    Royal Society Open Science , volume=

    Value attributed to text-based archives generated by artificial intelligence , author=. Royal Society Open Science , volume=. 2023 , url=

    work page 2023

  50. [69]

    Image Inspired Poetry Generation in XiaoIce

    Image inspired poetry generation in xiaoice , author=. arXiv preprint arXiv:1808.03090 , url=

    work page internal anchor Pith review Pith/arXiv arXiv

  51. [70]

    ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) , volume=

    Image to modern chinese poetry creation via a constrained topic-aware model , author=. ACM Transactions on Multimedia Computing, Communications, and Applications (TOMM) , volume=. 2020 , url=

    work page 2020

  52. [71]

    International Conference on Database Systems for Advanced Applications , pages=

    Poetic expression through scenery: sentimental Chinese classical poetry generation from images , author=. International Conference on Database Systems for Advanced Applications , pages=. 2021 , url=

    work page 2021

  53. [72]

    Proceedings of the 2023 3rd International Conference on Big Data, Artificial Intelligence and Risk Management , pages=

    Generating Chinese Poetry from Images Based on Deep Learning , author=. Proceedings of the 2023 3rd International Conference on Big Data, Artificial Intelligence and Risk Management , pages=

    work page 2023

  54. [73]

    Proceedings of the 26th ACM international conference on Multimedia , pages=

    Beyond narrative description: Generating poetry from images by multi-adversarial training , author=. Proceedings of the 26th ACM international conference on Multimedia , pages=

    work page

  55. [74]

    Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining , pages=

    Through the eyes of a poet: Classical poetry recommendation with visual input on social media , author=. Proceedings of the 2019 IEEE/ACM International Conference on Advances in Social Networks Analysis and Mining , pages=

    work page 2019

  56. [75]

    Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    How images inspire poems: Generating classical Chinese poetry from images with memory networks , author=. Proceedings of the AAAI Conference on Artificial Intelligence , volume=

    work page

  57. [76]

    中国古代诗歌情景论研究 , author=

    work page

  58. [77]

    Study on the Theory of Scene and Emotion in Ancient Chinese Poetry , author=

    work page

  59. [78]

    江西社会科学 , number=

    论中国古典诗歌情景交融的意象美 , author=. 江西社会科学 , number=. 2003 , url=

    work page 2003

  60. [79]

    Jiangxi Social Sciences , number=

    On the Imagery Beauty of Scene-Emotion Blending in Classical Chinese Poetry , author=. Jiangxi Social Sciences , number=. 2003 , url=

    work page 2003

  61. [80]

    百色学院学报 , number=

    中国古代早期诗歌创作中借景抒情手法形成的三个阶段 , author=. 百色学院学报 , number=

    work page

  62. [81]

    Journal of Baise University , number=

    The Three Stages in the Formation of the Technique of Expressing Emotions through Describing Scenes in Early Ancient Chinese Poetry Creation , author=. Journal of Baise University , number=. 2003 , url=

    work page 2003

  63. [82]

    淮南师范学院学报 , number=

    春恨秋怨何其多——中国古代诗歌作家创作心态浅探 , author=. 淮南师范学院学报 , number=

    work page

  64. [83]

    Journal of Huainan Normal University , number=

    Why So Many Spring Sorrows and Autumn Regrets: A Preliminary Exploration of the Creative Mentality of Ancient Chinese Poets , author=. Journal of Huainan Normal University , number=. 2001 , url=

    work page 2001

  65. [84]

    arXiv preprint arXiv:2508.13152 , url=

    RepreGuard: Detecting LLM-Generated Text by Revealing Hidden Representation Patterns , author=. arXiv preprint arXiv:2508.13152 , url=

    work page arXiv

  66. [85]

    arXiv preprint arXiv:2405.04286 , url=

    Who wrote this? the key to zero-shot llm-generated text detection is gecscore , author=. arXiv preprint arXiv:2405.04286 , url=

    work page arXiv

  67. [86]

    文学评论 , number=

    困境与出路:对当前新诗的思考 , author=. 文学评论 , number=

    work page

  68. [87]

    Literary Review , number=

    Dilemma and Solution: Reflections on Current New Poetry , author=. Literary Review , number=

    work page

  69. [88]

    华中师范大学学报：人文社会科学版 , volume=

    何以"扭断语法的脖子"——现代诗语修辞研究之一 , author=. 华中师范大学学报：人文社会科学版 , volume=

    work page

  70. [89]

    twist the neck of grammar

    Why "twist the neck of grammar" - a study of modern poetic rhetoric , author=. Journal of Central China Normal University: Humanities and Social Sciences Edition , volume=

    work page

  71. [90]

    Poetics , volume=

    Sentence judgments and the grammar of poetry: Linking linguistic structure and poetic effect , author=. Poetics , volume=. 2018 , publisher=

    work page 2018

  72. [91]

    2023 IEEE 14th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) , pages=

    How to detect AI-generated texts? , author=. 2023 IEEE 14th Annual Ubiquitous Computing, Electronics & Mobile Communication Conference (UEMCON) , pages=. 2023 , url=

    work page 2023

  73. [92]

    Proceedings of 3rd International Conference on Document Analysis and Recognition , author=

    Random decision forests. Proceedings of 3rd International Conference on Document Analysis and Recognition , author=. 1995 , publisher=

    work page 1995

  74. [93]

    Cornell University , year=

    XGBoost: A Scalable Tree Boosting System , author=. Cornell University , year=

    work page

  75. [94]

    Computer Science

    Improving image generation with better captions , author=. Computer Science. , url=

    work page

  76. [95]

    Gpt-image-1 model documentation , author=

    work page

  77. [96]

    Gpt-4o technical report , author=

    work page

  78. [97]

    GPT-4o System Card

    Gpt-4o system card , author=. arXiv preprint arXiv:2410.21276 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  79. [98]

    Qwen-Image Technical Report

    Qwen-image technical report , author=. arXiv preprint arXiv:2508.02324 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

  80. [99]

    HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer

    HiDream-I1: A High-Efficient Image Generative Foundation Model with Sparse Diffusion Transformer , author=. arXiv preprint arXiv:2505.22705 , year=

    work page internal anchor Pith review Pith/arXiv arXiv

Showing first 80 references.