浏览全部资源
扫码关注微信
Received：13 April 2026，
Revised：2026-07-22，
Accepted：10 August 2026，
Online First：12 August 2026，
移动端阅览
目的
2
图像辅助词汇习得是国际中文教育的重要教学策略，但现有提示词生成方法未能有效融合国际中文教育等级标准等领域专业知识，导致所生成图像在词义指向性、认知难度匹配性与教学适用性等方面存在显著不足。针对上述问题，本文提出一种面向国际中文教育词汇图像生成的渐进式提示词扩展模型（ Progressive prompt expansion model for vocabulary image generation，PPEM-VIG）。
方法
2
PPEM-VIG将词汇的认知等级、难度等级、词性、释义等离散教学要素统一建模为结构化教学要素集，并通过渐进式两阶段优化策略自动生成语义准确、结构规范且教学适用的图像提示词。监督学习阶段基于LLaMA模型以模板结构一致性损失与语言流畅性损失为约束，实现提示词的结构化生成；强化学习阶段基于监督微调LLaMA模型并增加模板内容相关性与基于CLIP的词图语义一致性奖励，对提示词进行深度优化，以实现提示词的拓展生成。
结果
2
本文构建了首个面向国际中文教育场景的词汇图像提示词数据集（international chinese language education vocabulary image prompt dataset， ICLE-VIPD）。实验结果表明，PPEM-VIG在模板结构覆盖率、CLIP语义一致性得分、及教师主观评价等核心指标上较最优的基线模型分别提升了4.10%，3.13%，4.54%。
结论
2
本文提出的渐进式监督-强化优化范式为词汇教学图像资源智能化生成上提供可借鉴的技术路线。
Objective
2
Image-assisted vocabulary acquisition has long been regarded as an effective pedagogical strategy in international Chinese language education， especially for helping learners establish intuitive connections between lexical meaning， visual representation， and communicative context. Compared with purely verbal explanation， instructional images can reduce cognitive load， support dual coding， and enhance learners’ retention of Chinese vocabulary. This is particularly important because Chinese words often involve polysemy， contextual dependence， cultural connotation， and different levels of visual expressibility. With the rapid development of text-to-image generation models， such as DALL-E， Stable Diffusion， and Midjourney， teachers are now able to generate customized visual materials at a lower cost and with greater flexibility. However， current prompt generation and prompt expansion methods are mainly designed for general-purpose image generation. They usually emphasize visual details， artistic style， lighting， composition， or aesthetic quality， but pay insufficient attention to pedagogical constraints in international Chinese language education. As a result， images generated from ordinary prompts may fail to accurately represent the target word meaning， may not match learners’ cognitive level or vocabulary difficulty， and may even introduce irrelevant or misleading visual elements. Existing methods also lack an effective mechanism for integrating domain-specific knowledge， such as vocabulary difficulty levels， cognitive stages， parts of speech， semantic explanations， and the Chinese proficiency grading standards for international Chinese language education. To address these limitations， this paper proposes a progressive prompt expansion model for vocabulary image generation， named PPEM-VIG， which aims to generate semantically accurate， structurally standardized， and pedagogically appropriate image prompts for Chinese vocabulary teaching.
Method
2
The proposed PPEM-VIG model formulates vocabulary image prompt generation as a structured mapping problem from pedagogical elements to image-generation prompts. Specifically， discrete teaching-related information， including cognitive grade， difficulty level， part of speech， and semantic explanation， is organized into a structured pedagogical element set. This representation enables the model to explicitly capture the instructional requirements of each vocabulary item rather than treating the word as an isolated text input. On this basis， PPEM-VIG adopts a progressive two-stage optimization framework. In the first stage， supervised learning is used to train a LLaMA-based structured prompt generation model. The model learns the mapping from the pedagogical element set to a structured prompt template consisting of several functional blocks， such as scene description， core vocabulary representation， primary visual features， auxiliary expression， pedagogical purpose， and style or cultural adaptation. To ensure that the generated prompt follows the expected template and remains readable， this stage introduces template structure consistency loss and language fluency loss. The former constrains the completeness and order of prompt blocks， while the latter encourages natural and coherent language generation.In the second stage， reinforcement learning is introduced to further optimize the supervised fine-tuned model. Unlike conventional prompt expansion methods that focus mainly on textual enrichment， this stage incorporates multiple reward signals related to pedagogical relevance and cross-modal alignment. The reward function includes template structure consistency， language fluency， template content relevance， and CLIP-based text-image semantic consistency. Among these rewards， the content relevance reward evaluates whether the generated prompt accurately reflects the input vocabulary and its teaching elements， while the CLIP-based reward measures whether the image generated from the prompt is semantically consistent with the vocabulary definition. Through this multi-objective reinforcement learning process， PPEM-VIG can refine prompts beyond surface-level template completion and improve their ability to guide text-to-image models toward educationally meaningful visual outputs. The progressive combination of supervised learning and reinforcement learning allows the model to first acquire stable structural generation ability and then enhance semantic alignment and visual teaching applicability.
Result
2
To support model training and evaluation， this study constructs the first vocabulary image prompt dataset specifically designed for international Chinese language education， named ICLE-VIPD. The dataset is built around vocabulary items selected from the Chinese proficiency grading standards for international Chinese language education and focuses on high-frequency nouns and verbs across different difficulty levels. Each sample contains a pedagogical element set and a corresponding structured prompt. The prompt annotation process considers the different visual representation principles of nouns and verbs. For nouns， the prompt design emphasizes core visual features， typical scenes， contrastive elements， and cultural appropriateness. For verbs， the prompt design focuses on action participants， action process， affected objects， result states， and contextual cues. Expert annotators with backgrounds in international Chinese language teaching participate in the construction and verification process to ensure that the prompts are both semantically accurate and suitable for classroom use.Experiments are conducted to evaluate PPEM-VIG from multiple perspectives， including template structure coverage， language fluency， block relevance， CLIP semantic consistency， and teachers’ subjective evaluation. The results show that PPEM-VIG achieves superior performance compared with representative baseline models. In particular， compared with the strongest baseline model， PPEM-VIG improves template structure coverage by 4.10%， CLIP semantic consistency score by 3.13%， and teachers’ subjective evaluation score by 4.54%. These improvements indicate that the proposed model not only follows the structured prompt template more reliably but also generates prompts that better support text-to-image models in producing semantically aligned teaching images. The teacher evaluation results further confirm that PPEM-VIG-generated prompts are more effective in terms of word-meaning accuracy， teaching adaptability， clarity of expression， and completeness of instructional information. Ablation analysis also demonstrates the necessity of the two-stage framework. The supervised learning stage substantially improves structural completeness and pedagogical element coverage， while the reinforcement learning stage further enhances cross-modal semantic consistency and practical teaching value. Removing individual reward components leads to performance degradation， which verifies the contribution of each reward signal to the overall model.
Conclusion
2
The proposes PPEM-VIG， a progressive prompt expansion model for vocabulary image generation in international Chinese language education. By incorporating structured pedagogical elements and combining supervised learning with reinforcement learning， the model effectively addresses the limitations of existing general-purpose prompt generation methods in educational scenarios. PPEM-VIG can generate prompts that are not only visually descriptive but also aligned with vocabulary meaning， learner cognition， difficulty level， and teaching objectives. The construction of ICLE-VIPD further provides a valuable benchmark resource for future research on vocabulary-oriented image generation， prompt engineering， and intelligent teaching material generation. Overall， the proposed progressive supervised-reinforcement optimization paradigm offers a practical technical route for the intelligent generation of vocabulary teaching images and contributes to the digital and intelligent transformation of international Chinese language education.
Song F and Abdul Rabu S N . 2025 . Trends， advantages， and challenges： a systematic literature review of artificial intelligence in design education . Journal of Educational and Social Research ， 15 （ 4 ）： 401 - 417 ［ DOI： 10.36941/jesr-2025-0147 http://dx.doi.org/10.36941/jesr-2025-0147 ］
Zhao X T . 2021 . On the important position of vocabulary teaching in international Chinese teaching . Xin Jishi ， （ 9 ）： 45 - 47
赵雪婷 . 2021 . 试论词汇教学在国际中文教学中的重要地位 . 新纪实 ， （ 9 ）： 45 - 47
Rao J . 2021 . Effects of images and subtitles on incidental vocabulary acquisition of eighth-grade students under audio-visual input . Changsha ： Hunan University
饶静 . 2021 . 视听输入下图像和字幕对初二学生词汇附带习得的影响 . 长沙 ： 湖南大学
Zheng Y Q and Chen W H . 2006 . A study on picture expression methods of HSK nouns . Chinese Teaching in the World ， （ 4 ）： 107 - 115
郑艳群 ， 陈文慧 . 2006 . HSK名词图片表达方法研究 . 世界汉语教学 ， （ 4 ）： 107 - 115 ［ DOI： 10.13724/j.cnki.ctiw.2006.04.016 http://dx.doi.org/10.13724/j.cnki.ctiw.2006.04.016 ］
Lu N . 2009 . Analysis on the picture expressibility of HSK verbs from semantic features . Beijing ： Beijing Language and Culture University
卢娜 . 2009 . 从语义特征分析HSK动词的图片可表达性 . 北京 ： 北京语言大学
Wang J . 2022 . Alignment and adaptation： vocabulary teaching strategies based on the Chinese Proficiency Grading Standards for International Chinese Language Education . Journal of International Chinese Teaching ， （ 4 ）： 10 - 19
王军 . 2022 . 对接与调适：基于《国际中文教育中文水平等级标准》的词汇教学策略 . 国际汉语教学研究 ， （ 4 ）： 10 - 19
Bharne S ， Sapkale P ， Sarda E ， Salunkhe S and Padiya P . 2025 . Towards sustainable image synthesis： a comprehensive review of text-to-image generation models . International Research Journal of Multidisciplinary Technovation ， 7 （ 5 ）： 94 - 120 ［ DOI： 10.54392/irjmt2557 http://dx.doi.org/10.54392/irjmt2557 ］
Kadry K ， Gupta S ， Nezami F R and Edelman E R . 2024 . Probing the limits and capabilities of diffusion models for the anatomic editing of digital twins . npj Digital Medicine ， 7 （ 1 ） ［ DOI： 10.1038/s41746-024-01332-0 http://dx.doi.org/10.1038/s41746-024-01332-0 ］
Cao Y . 2024 . Study on the utilization of pictures in vocabulary instruction for international Chinese language education . Journal of Education and Educational Research ， 9 （ 1 ）： 19 - 22 ［ DOI： 10.54097/dbkgdg2 http://dx.doi.org/10.54097/dbkgdg2
Sudha L ， Aruna K B ， Sureka V ， Niveditha M and Prema S . 2024 . Semantic image synthesis from text： current trends and future horizons in text-to-image generation . EAI Endorsed Transactions on Internet of Things ， 11 ［ DOI： 10.4108/eetiot.5336 http://dx.doi.org/10.4108/eetiot.5336 ］
Purwono P ， Wulandari A N E ， Ma’arif A and Salah W A . 2025 . Understanding generative adversarial networks： a review . Control Systems and Optimization Letters ， 3 （ 1 ）： 36 - 45 ［ DOI： 10.59247/csol.v3i1.170 http://dx.doi.org/10.59247/csol.v3i1.170 ］
Ji Y R ， Wang C H ， Chen J B ， Yue A Z ， Xi Z H and Chen J S . Parameter-efficient diffusion model adaptation and spectral consistency learning for controllable multispectral remote sensing image generation ［J/OL］. Journal of Image and Graphics ： 1 - 16
纪璎芮 ， 王晨昊 ， 陈静波 ， 岳安志 ， 席智浩 ， 陈建胜 . 多光谱遥感图像可控生成的扩散模型参数高效适配与光谱一致性学习 ［J/OL］. 中国图象图形学报 ： 1 - 16 ［ DOI： 10.11834/jig.260089 http://dx.doi.org/10.11834/jig.260089 ］
Shafi S . 2024 . Text-to-image generation using stack generative adversarial networks and stable diffusion models . International Journal for Research in Applied Science and Engineering Technology ， 12 （ 11 ）： 1426 - 1429 ［ DOI： 10.22214/ijraset.2024.65350 http://dx.doi.org/10.22214/ijraset.2024.65350 ］
Wang R . 2024 . A comparative analysis of StackGAN and AttnGAN in text-to-image generation . Applied and Computational Engineering ， 105 （ 1 ）： 9 - 15 ［ DOI： 10.54254/2755-2721/105/2024tj0055 http://dx.doi.org/10.54254/2755-2721/105/2024tj0055 ］
Ho J ， Jain A and Abbeel P . 2020 . Denoising diffusion probabilistic models ［EB/OL］. ［ DOI： 10.48550/ARXIV.2006.11239 http://dx.doi.org/10.48550/ARXIV.2006.11239 ］
Ramesh A ， Dhariwal P ， Nichol A ， Chu C and Chen M . 2022 . Hierarchical text-conditional image generation with CLIP latents ［EB/OL］. ［ DOI： 10.48550/ARXIV.2204.06125 http://dx.doi.org/10.48550/ARXIV.2204.06125 ］
Saharia C ， Chan W ， Saxena S ， Li L ， Whang J ， Denton E ， et al . 2022 . Photorealistic text-to-image diffusion models with deep languageunderstanding ［EB/OL］.［ DOI： 10.48550/ARXIV.2205.11487 http://dx.doi.org/10.48550/ARXIV.2205.11487 ］
Rombach R ， Blattmann A ， Lorenz D ， Esser P and Ommer B . 2021 . High-resolution image synthesis with latent diffusion models ［EB/OL］. ［ DOI： 10.48550/ARXIV.2112.10752 http://dx.doi.org/10.48550/ARXIV.2112.10752 ］
Schneider J . 2024 . Explainable generative AI： a survey， conceptualization， and research agenda . Artificial Intelligence Review ， 57 （ 11 ） ［ DOI： 10.1007/s10462-024-10916-x http://dx.doi.org/10.1007/s10462-024-10916-x ］
Yang L ， Zhang Z ， Song Y ， Hong S ， Xu R ， Zhao Y ， et al . 2023 . Diffusion models： a comprehensive survey of methods and applications . ACM Computing Surveys ， 56 （ 4 ）： 1 - 39 ［ DOI： 10.1145/3626235 http://dx.doi.org/10.1145/3626235 ］
Chen Z ， Zhang L ， Weng F ， Pan L and Lan Z . 2024 . Tailored visions： enhancing text-to-image generation with personalized prompt rewriting // Proceedings of the 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition . Seattle， USA ： IEEE/CVF： 7727 - 7736 ［ DOI： 10.1109/CVPR52733.2024.00792 http://dx.doi.org/10.1109/CVPR52733.2024.00792 ］
Liu V and Chilton L B . 2022 . Design guidelines for prompt engineering text-to-image generative models // Proceedings of the 2022 CHI Conference on Human Factors in Computing Systems . New Orleans，USA ： ACM： 1 - 23 ［ DOI： 10.1145/3491102.3501825 http://dx.doi.org/10.1145/3491102.3501825 ］
Cao T ， Wang C ， Liu B ， et al . 2023 . Beautifulprompt： towards automatic prompt engineering for text-to-image synthesis // Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing： Industry Track . Singapore ： Association for Computational Linguistics： 1 - 11 ［ DOI： 10.18653/v1/2023.emnlp-industry.1 http://dx.doi.org/10.18653/v1/2023.emnlp-industry.1 ］
Lu Y H ， Feng Y C J ， Zhu L ， Zhou H Y ， Zhu H ， Yu C H ， et al . 2025 . PromptVis： prompt-based interactive visual analysis method for text-to-image creation . Journal of Computer-Aided Design & Computer Graphics ， 37 （ 4 ）： 688 - 696
卢裕弘 ， 封颖超杰 ， 朱琳 ， 周海怡 ， 朱航 ， 喻晨昊 ， 等 . 2025 . PromptVis ： 面向文本生成图片的提示词的交互式可视分析方法. 计算机辅助设计与图形学学报） ， 37 （ 4 ）： 688 - 696 ［ DOI： 10.3724/SP.J.1089.2023-00344 http://dx.doi.org/10.3724/SP.J.1089.2023-00344 ］
Fernando DVA ， Atharva T ， Patil A ， Sinha D and Ludup T . 2025 . Harnessing stable diffusion model for high-resolution text-to-image synthesis . International Journal of Scientific Research in Engineering and Management ， 9 （ 5 ）： 1 - 9 ［ DOI： 10.55041/ijsrem48372 http://dx.doi.org/10.55041/ijsrem48372 ］
Kitrungrotsakul T ， Xu Y and Srichola P . 2025 . Knowledge distillation meets reinforcement learning： a cluster-driven approach to image processing . Sensors ， 26 （ 1 ）： 209 ［ DOI： 10.3390/s26010209 http://dx.doi.org/10.3390/s26010209 ］
Datta S ， Ku A ， Ramachandran D ， et al . 2024 . Prompt expansion for adaptive text-to-image generation // Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics . Bangkok， Thailand ： Association for Computational Linguistics： 3449 - 3476 ［ DOI： 10.18653/v1/2024.acl-long.189 http://dx.doi.org/10.18653/v1/2024.acl-long.189 ］
Leong H H M ， Chen Q ， Ye Z ， et al . 2025 . PromptBoost ： annotation-free prompt expansion for Chinese text-to-image generation// Proceedings of the 3rd International Workshop on Large Generative Models Meet Multimodal Applications ： 3 - 10 ［ DOI： 10.1145/3728422.3762144 http://dx.doi.org/10.1145/3728422.3762144 ］
Li C ， Zhang M ， Mei Q ， Kong W and Bendersky M . 2024 . Learning to rewrite prompts for personalized text generation // Proceedings of the ACM Web Conference 2024 . Singapore ： ACM： 3367 - 3378 ［ DOI： 10.1145/3589334.3645408 http://dx.doi.org/10.1145/3589334.3645408 ］
Nori H ， Lee Y T ， Zhang S ， Carignan D ， Edgar R ， Fusi N ， et al . 2023. Can generalist foundation models outcompete special-purpose tuning？ Case study in medicine ［EB/OL］. ［ DOI： 10.48550/arXiv.2311.16452 http://dx.doi.org/10.48550/arXiv.2311.16452 ］
Liu B ， Wang L ， Lyu C ， Zhang Y ， Su J ， Shi S ， et al . 2023 . On the cultural gap in text-to-image generation ［EB/OL］. ［ DOI： 10.48550/arXiv.2307.02971 http://dx.doi.org/10.48550/arXiv.2307.02971 ］
Ahmadi M ， Zarei A A and Esfandiari R . 2020 . Learning L2 idioms through visual mnemonics . Journal of English Language Teaching and Learning ， 12 （ 26 ）： 1 - 27 ［ DOI： 10.22034/elt.2020.11438 http://dx.doi.org/10.22034/elt.2020.11438 ］
Choi W C and Chang C I . 2025 . A survey of techniques， key components， strategies， challenges， and student perspectives on prompt engineering for large language models in education ［EB/OL］. ［ DOI： 10.20944/preprints202503.1808.v1 http://dx.doi.org/10.20944/preprints202503.1808.v1 ］
Shi J ， Qi M ， Zhang L ， Wang D ， Zhao Y ， Li Z ， et al . 2025 . Collaborative text-to-image generation via multi-agent reinforcement learning and semantic fusion ［EB/OL］. ［ DOI： 10.48550/ARXIV.2510.10633 http://dx.doi.org/10.48550/ARXIV.2510.10633 ］
Ramesh A ， Pavlov M ， Goh G ， Gray S ， Voss C ， Radford A ， et al . 2021 . Zero-shot text-to-image generation ［EB/OL］. ［ DOI： 10.48550/ARXIV.2102.12092 http://dx.doi.org/10.48550/ARXIV.2102.12092 ］
Qi Y and Guo D . 2024 . Enhancing text-to-image generation with diversity regularization and fine-grained supervision . Highlights in Science ， Engineering and Technology ， 122 ： 1 - 9 ［ DOI： 10.54097/42m6by18 http://dx.doi.org/10.54097/42m6by18 ］
Mayer R E . 2005 . The Cambridge handbook of multimedia learning . Cambridge ： Cambridge University Press
Mayer R E . 2005 . 多媒体学习剑桥手册 . 剑桥 ： 剑桥大学出版社
Paivio A . 1990 . Mental representations： a dual coding approach . Oxford ： Oxford University Press
Paivio A . 1990 . 心理表征：双重编码法 . 牛津 ： 牛津大学出版社
Shen H H . 2010 . Imagery and verbal coding approaches in Chinese vocabulary instruction . Language Teaching Research ， 14 （ 4 ）： 485 - 499 ［ DOI： 10.1177/1362168810375370 http://dx.doi.org/10.1177/1362168810375370 ］
Li J T and Tong F . 2020 . The effect of cognitive vocabulary learning approaches on Chinese learners’ compound word attainment， retention， and learning motivation . Language Teaching Research ， 24 （ 6 ）： 834 - 854 ［ DOI： 10.1177/1362168819829923 http://dx.doi.org/10.1177/1362168819829923 ］
Xiong T and Peng Y . 2021 . Representing culture in Chinese as a second language textbooks： a critical social semiotic approach . Language， Culture and Curriculum ， 34 （ 2 ）： 163 - 182 ［ DOI： 10.1080/07908318.2020.1797079 http://dx.doi.org/10.1080/07908318.2020.1797079 ］
Gou D . 2025 . The potential， challenges， and pathways of generative artificial intelligence in empowering the professional development of international Chinese language teachers . Journal of Current Social Issues Studies ， 2 （ 2 ）： 104 - 115 ［ DOI： 10.56397/JCSIS.2025.02.13 http://dx.doi.org/10.56397/JCSIS.2025.02.13 ］
Yu J ， Song J and Lu Y . 2025 . Harnessing generative artificial intelligence to construct multimodal resources for Chinese character learning . Systems ， 13 （ 8 ）： 692 ［ DOI： 10.3390/systems13080692 http://dx.doi.org/10.3390/systems13080692 ］
Attygalle N T ， Kljun M ， Quigley A ， et al . 2025 . Text-to-image generation for vocabulary learning using the keyword method // Proceedings of the 30th International Conference on Intelligent User Interfaces . Cagliari， Italy ： ACM： 1381 - 1397 ［ DOI： 10.1145/3670653.3703358 http://dx.doi.org/10.1145/3670653.3703358 ］
Fan K . 2025 . Generative AI and second language vocabulary processing： a cognitive study of Chinese EFL learners . Journal of Humanities and Social Sciences Studies ， 7 （ 7 ）： 103 - 111 ［ DOI： 10.32996/jhsss.2025.7.7.12 http://dx.doi.org/10.32996/jhsss.2025.7.7.12 ］
Shi X C . 2020 . Research on Chinese description generation technology for automobile evaluation images . Guangzhou ： Guangdong University of Technology
史秀聪 . 2020 . 汽车评测图像中文描述生成技术研究 . 广州 ： 广东工业大学
Yang C L ， Wan W G ， Wang X Z ， Sun X T and Zhang Z . 2023 . Image semantic understanding based on text perception and non-repeating word generation . Industrial Control Computer ， 11 ： 105 - 106， 109
杨晨露 ， 万旺根 ， 王旭智 ， 孙学涛 ， 张振 . 2023 . 基于文本感知和非重复单词生成的图像语义理解 . 工业控制计算机 ， 11 ： 105 - 106 ， 109
Zhao X W ， Zhu Z T and Shen S S . 2023 . Prompt engineering in education： constructing an epistemological new discourse in the era of digital intelligence . China Distance Education ， 43 （ 11 ）： 22 - 31
赵晓伟 ， 祝智庭 ， 沈书生 . 2023 . 教育提示语工程：构建数智时代的认识论新话语 . 中国远程教育 ， 43 （ 11 ）： 22 - 31 ［ DOI： 10.13541/j.cnki.chinade.2023.11.003 http://dx.doi.org/10.13541/j.cnki.chinade.2023.11.003 ］
Touvron H ， Lavril T ， Izacard G ， Martinet X ， Lachaux M A ， Lacroix T ， et al . 2023 . LLaMA： open and efficient foundation language models ［EB/OL］.［ 2026-05-17 ］.［ DOI： 10.48550/arXiv.2302.13971 http://dx.doi.org/10.48550/arXiv.2302.13971 ］
Brown T B ， Mann B ， Ryder N ， Subbiah M ， Kaplan J D ， Dhariwal P ， et al . 2020 . Language models are few-shot learners // Proceedings of the 34th International Conference on Neural Information Processing Systems . Virtual ： Curran Associates， Inc.： 1877 - 1901
Taori R ， Gulrajani I ， Zhang T ， Dubois Y ， Li X ， Guestrin C ， et al . 2023 . Stanford alpaca： an instruction-following LLaMA model ［EB/OL］.［ 2026-05-17 ］. https://github.com/tatsu-lab/stanford_alpaca https://github.com/tatsu-lab/stanford_alpaca
Hu E J ， Shen Y ， Wallis P ， Allen-Zhu Z ， Li Y ， Wang S ， et al . 2022 . LoRA： low-rank adaptation of large language models // Proceedings of the 10th International Conference on Learning Representations . Virtual ： OpenReview.net
Ouyang L ， Wu J ， Jiang X ， Almeida D ， Wainwright C ， Mishkin P ， et al . 2022 . Training language models to follow instructions with human feedback // Proceedings of the 36th International Conference on Neural Information Processing Systems . New Orleans， USA ： Curran Associates， Inc.： 27730 - 27744
Chen S Q ， Yang X ， Zhu R Q ， Liao N and Zhao W W . 2026 . Parameter-efficient fine-tuning for remote sensing image interpretation： a survey . Journal of Image and Graphics ， 31 （ 1 ）： 0212 - 0242
陈诗琪 ， 杨学 ， 朱荣强 ， 廖宁 ， 赵卫伟 . 2026 . 面向遥感图像解译的参数高效微调研究综述 . 中国图象图形学报 ， 31 （ 1 ）： 0212 - 0242 ［ DOI： 10.11834/jig.250105 http://dx.doi.org/10.11834/jig.250105 ］
Center for Language Education and Cooperation . 2025 . Chinese graded readers standards for international Chinese language education . Beijing ： Beijing Language and Culture University Press （中外语言交流合作中心） . 2025. 国际中文教育中文阅读分级标准 . 北京： 北京语言大学出版社）
相关文章
相关作者
相关机构
京公网安备11010802024621
