@article{krishnaApplyingTextAnalytics2021,
	title = {Applying {Text} {Analytics} to the {Mind}-section {Literature} of the {Tibetan} {Tradition} of the {Great} {Perfection}},
	volume = {20},
	issn = {2375-4699, 2375-4702},
	url = {https://dl.acm.org/doi/10.1145/3392047},
	doi = {10.1145/3392047},
	abstract = {Over the past decade, through a mixture of optical character recognition and manual input, there is now a growing corpus of Tibetan literature available as e-texts in Unicode format. With the creation of such a corpus, the techniques of text analytics that have been applied in the analysis of English and other modern languages may now be applied to Tibetan. In this work, we narrow our focus to examine a modest portion of that literature, the Mind-section portion of the literature of the Tibetan tradition of the Great Perfection. Here, we will use the lens of text analytics tools based on machine learning techniques to investigate a number of questions of interest to scholars of this and related traditions of the Great Perfection. It has been necessary for us to participate in all portions of this process: corpora identification and text edition selection, rendering the text as e-texts in Unicode using both Optical Character Recognition and manual entry, data cleaning and transformation, implementation of software for text analysis, and interpretation of results. For this reason, we hope this study can serve as a model for other low-resource languages that are just beginning to approach the problem of providing text analytics for their language.},
	language = {en},
	number = {2},
	urldate = {2026-02-09},
	journal = {ACM Transactions on Asian and Low-Resource Language Information Processing},
	author = {Krishna, Ravi and Mu, Norman and Keutzer, Kurt},
	month = mar,
	year = {2021},
	pages = {1--32},
	file = {Volltext:/home/thomas/Zotero/storage/MA7LMLZT/Krishna et al. - 2021 - Applying Text Analytics to the Mind-section Literature of the Tibetan Tradition of the Great Perfect.pdf:application/pdf},
}

@article{huangSunShineLargeLanguage2024,
	title = {Sun-{Shine}: {A} {Large} {Language} {Model} for {Tibetan} {Culture}},
	journal = {arXiv preprint arXiv:2407.10671},
	author = {Huang, Cheng and Gao, Fan and Tashi, Nyima and Liu, Yutong et al.},
	year = {2024},
}

@misc{huangTFDComprehensiveStructured2026,
	title = {{TFD}: {A} {Comprehensive} {Structured} {Tibetan} {Foundation} {Dataset} for {Low}-{Resource} {Language} {Processing} and {Large}-{Scale} {Modeling}},
	shorttitle = {{TFD}},
	url = {http://arxiv.org/abs/2503.18288},
	doi = {10.48550/arXiv.2503.18288},
	abstract = {Large language models (LLMs) have achieved remarkable success in high-resource languages, yet progress for Tibetan remains severely constrained by the lack of large-scale, high-quality, and structured data. Existing Tibetan resources are fragmented, domain-limited, and insufficient to support modern LLM pipelines requiring pretraining, instruction tuning, safety alignment, and reasoning supervision. We introduce the {\textbackslash}textbf\{T\}ibetan {\textbackslash}textbf\{F\}oundation {\textbackslash}textbf\{D\}ataset ({\textbackslash}textbf\{TFD\}), the first comprehensive, large-scale, and expert-curated dataset explicitly designed for Tibetan large language modeling. {\textbackslash}textit\{TFD\} comprises two complementary components: {\textbackslash}textit\{TIBSTC\}, a unified corpus of over 11 billion tokens spanning literature, law, medicine, religion, and everyday communication, and {\textbackslash}textit\{TIBSTC-CoT\}, the first large-scale Tibetan chain-of-thought dataset supporting explicit multi-step reasoning across diverse domains. Unlike prior Tibetan datasets, {\textbackslash}textit\{TFD\} is structurally organized to support the full LLM development lifecycle, including pretraining, supervised fine-tuning, safety alignment, and preference optimization. We demonstrate its utility by training the {\textbackslash}textit\{Sun-Shine\} family of Tibetan LLMs and evaluating them on understanding, safety, reasoning, and generation tasks. Results show consistent improvements over strong open-source and proprietary baselines, underscoring the importance of large-scale, structured data for low-resource language modeling. We release {\textbackslash}textit\{TFD\} to facilitate reproducible research and the development of robust, culturally aligned Tibetan LLMs. Code and data are available at https://github.com/Vicentvankor/sun-shine.},
	urldate = {2026-02-26},
	publisher = {arXiv},
	author = {Huang, Cheng and Gao, Fan and Tashi, Nyima and Liu, Yutong and Wang, Xiangxiang and Tsering, Thupten and Ma-bao, Ban and Feng, Xiao and Duojie, Renzeg and Luosang, Gadeng and Dongrub, Rinchen and Tashi, Dorje and Wang, Hao and Yu, Yongbin},
	month = feb,
	year = {2026},
	note = {arXiv:2503.18288 [cs]},
	keywords = {Computer Science - Computation and Language},
	file = {Preprint PDF:/home/thomas/Zotero/storage/QUIKUI4U/Huang et al. - 2026 - TFD A Comprehensive Structured Tibetan Foundation Dataset for Low-Resource Language Processing and.pdf:application/pdf;Snapshot:/home/thomas/Zotero/storage/VFUJGSAW/2503.html:text/html},
}

@article{liCharacterbasedJointWord2022,
	title = {Character-based {Joint} {Word} {Segmentation} and {Part}-of-{Speech} {Tagging} for {Tibetan} {Based} on {Deep} {Learning}},
	volume = {21},
	issn = {2375-4699, 2375-4702},
	url = {https://dl.acm.org/doi/10.1145/3511600},
	doi = {10.1145/3511600},
	abstract = {Tibetan word segmentation and POS tagging are the primary tasks of Tibetan natural language processing. Most of existing methods of Tibetan word segmentation and POS tagging are based on rules and statistics, which need manual construction of features. In addition, the joint mode has shown stronger capabilities for word segmentation and POS tagging and have received great interests. In this paper, we propose Bi-LSTM+IDCNN+CRF structures, a simple yet effective end-to-end neural network model, for joint Tibetan word segmentation and POS tagging. We conduct step-by-step and joint experiments on the Tibetan datasets. The results demonstrate that the performance of the Bi-LSTM+IDCNN+CRF model is the best regardless of the step-by-step or joint mode. We obtain state-of-the-art performance in the joint tagging mode. The F1 score of the word segmentation task reached 92.31\%, and the F1 score of the POS tagging task reached 81.26\%.},
	language = {en},
	number = {5},
	urldate = {2026-02-26},
	journal = {ACM Transactions on Asian and Low-Resource Language Information Processing},
	author = {Li, Yan and Li, Xiaomin and Wang, Yiru and Lv, Hui and Li, Fenfang and Duo, La},
	month = sep,
	year = {2022},
	pages = {1--15},
}

@misc{meelenAnnotatedCorpusClassical2020,
	title = {The {Annotated} {Corpus} of {Classical} {Tibetan} ({ACTib}) - {Version} 2.0 ({Segmented} \& {POS}-tagged)},
	copyright = {Creative Commons Attribution 4.0 International},
	url = {https://zenodo.org/doi/10.5281/zenodo.3951503},
	doi = {10.5281/ZENODO.3951503},
	abstract = {This corpus consisting of {\textgreater}185 million tokens is a segmented and part-of-speech tagged version of

Wallman, Jeff, Rowinski, Zach, Ngawang Trinley, Tomlinson, Chris, \& Keutzer, Kurt. (2017). Collection of Tibetan etexts compiled by the Buddhist Digital Resource Center [Data set]. Zenodo. http://doi.org/10.5281/zenodo.821218

using the training data of

Hill, Nathan W., \& Garrett, Edward. (2017). A part-of-speech (POS) tagged corpus of Classical Tibetan [Data set]. Zenodo. http://doi.org/10.5281/zenodo.574878

The code for segmenting and POS tagging any Tibetan file can be found on GitHub.

This Version 2 of ACTib is based on the same XML files as ACTib Version 1 (http://doi.org/10.5281/zenodo.823707), but contains both segmented and POS-tagged files and is improved in a number of ways, although post-processing was still done automatically and no manual correction was involved. For details of this improved annotation method see:

Meelen, Marieke, Roux, Élie \& Hill, Nathan (2021). 'Optimisation of the largest annotated Tibetan corpus combining rule-based, memory-based \& deep-learning methods' in ACM Transactions on Asian and Low-Resource Language Information Processing (TALLIP) , 20 , pp. 1–11.},
	urldate = {2026-02-26},
	publisher = {Zenodo},
	author = {Meelen, Marieke and Roux, Élie},
	month = may,
	year = {2020},
	keywords = {Annotated Historical Corpus, Natural Language Processing, POS tagging, Segmentation, Tibetan language},
}

@article{meelenOptimisationLargestAnnotated2021,
	address = {New York, NY, USA},
	title = {Optimisation of the {Largest} {Annotated} {Tibetan} {Corpus} {Combining} {Rule}-based, {Memory}-based, and {Deep}-learning {Methods}},
	volume = {20},
	issn = {2375-4699},
	url = {https://doi.org/10.1145/3409488},
	doi = {10.1145/3409488},
	abstract = {This article presents a pipeline that converts collections of Tibetan documents in plain text or XML into a fully segmented and POS-tagged corpus. We apply the pipeline to the large extent collection of the Buddhist Digital Resource Center. The semi-supervised methods presented here not only result in a new and improved version of the largest annotated Tibetan corpus to date, the integration of rule-based, memory-based, and neural-network methods also serves as a good example of how to overcome challenges of under-researched languages. The end-to-end accuracy of our entire automatic pipeline of 91.99\% is high enough to make the resulting corpus a useful resource for both linguists and scholars of Tibetan studies.},
	number = {1},
	journal = {ACM Trans. Asian Low-Resour. Lang. Inf. Process.},
	publisher = {Association for Computing Machinery},
	author = {Meelen, Marieke and Roux, Élie and Hill, Nathan},
	month = mar,
	year = {2021},
	keywords = {historical treebanks, NLP, POS tagging, Tibetan},
}

@article{erhardTextInformationExtraction2025,
	title = {Text and {Information} {Extraction} for {Modern} (post-1950) {Tibetan} {Newspapers} and {Print} {Publications} with {Transkribus}},
	number = {74},
	journal = {Revue d’Etudes Tibétaines},
	author = {Erhard, Franz Xaver},
	month = feb,
	year = {2025},
	pages = {128--171},
}

@misc{panAdvancingLargeLanguage2025,
	title = {Advancing {Large} {Language} {Models} for {Tibetan} with {Curated} {Data} and {Continual} {Pre}-{Training}},
	url = {http://arxiv.org/abs/2507.09205},
	doi = {10.48550/arXiv.2507.09205},
	abstract = {Large language models have achieved remarkable progress across many languages. However, Tibetan, as a representative low-resource language, is particularly underrepresented in existing models due to the scarcity of high-quality training corpora. To address this gap, we curate the largest Tibetan pre-training corpus to date, aggregating data from diverse sources and applying a dedicated data cleaning and processing pipeline tailored for Tibetan. With the curated data, we continue pre/post-training a multilingual base model to enhance its generative capabilities in Tibetan. To evaluate the Tibetan capabilities of the model, we create new high-quality Tibetan benchmarks, and complement them with existing public benchmarks. Experimental results demonstrate that our model consistently and significantly outperforms both open-source models of similar scale and Tibetan-tailored models across a wide range of tasks.},
	urldate = {2026-02-26},
	publisher = {arXiv},
	author = {Pan, Leiyu and Xiong, Bojian and Yang, Lei and Jin, Renren and Zhang, Shaowei and Chen, Yue and Shi, Ling and Zhou, Jiang and Wu, Junru and Wang, Zhen and Peng, Jianxiang and Xiao, Juesi and Dong, Tianyu and Han, Zhuowen and Chen, Zhuo and Ren, Yuqi and Xiong, Deyi},
	month = jul,
	year = {2025},
	note = {arXiv:2507.09205 [cs]},
	keywords = {Computer Science - Computation and Language},
	file = {Preprint PDF:/home/thomas/Zotero/storage/9L2J2E8J/Pan et al. - 2025 - Advancing Large Language Models for Tibetan with Curated Data and Continual Pre-Training.pdf:application/pdf;Snapshot:/home/thomas/Zotero/storage/XBXRQ8P2/2507.html:text/html},
}

@article{meelenSegmentingPOSTagging2018,
	title = {Segmenting and {POS} tagging {Classical} {Tibetan} using a memory-based tagger},
	volume = {16},
	issn = {1544-7502},
	url = {https://escholarship.org/uc/item/8b83z79n},
	doi = {10.5070/H916234501},
	number = {2},
	urldate = {2026-02-26},
	journal = {Himalayan Linguistics},
	author = {Meelen, Marieke and Hill, Nathan},
	month = jan,
	year = {2018},
	file = {Volltext:/home/thomas/Zotero/storage/NDQKPZVR/Meelen und Hill - 2018 - Segmenting and POS tagging Classical Tibetan using a memory-based tagger.pdf:application/pdf},
}

@misc{huangTibetanLanguageAI2025,
	title = {Tibetan {Language} and {AI}: {A} {Comprehensive} {Survey} of {Resources}, {Methods} and {Challenges}},
	copyright = {Creative Commons Attribution 4.0 International},
	shorttitle = {Tibetan {Language} and {AI}},
	url = {https://arxiv.org/abs/2510.19144},
	doi = {10.48550/ARXIV.2510.19144},
	abstract = {Tibetan, one of the major low-resource languages in Asia, presents unique linguistic and sociocultural characteristics that pose both challenges and opportunities for AI research. Despite increasing interest in developing AI systems for underrepresented languages, Tibetan has received limited attention due to a lack of accessible data resources, standardized benchmarks, and dedicated tools. This paper provides a comprehensive survey of the current state of Tibetan AI in the AI domain, covering textual and speech data resources, NLP tasks, machine translation, speech recognition, and recent developments in LLMs. We systematically categorize existing datasets and tools, evaluate methods used across different tasks, and compare performance where possible. We also identify persistent bottlenecks such as data sparsity, orthographic variation, and the lack of unified evaluation metrics. Additionally, we discuss the potential of cross-lingual transfer, multi-modal learning, and community-driven resource creation. This survey aims to serve as a foundational reference for future work on Tibetan AI research and encourages collaborative efforts to build an inclusive and sustainable AI ecosystem for low-resource languages.},
	urldate = {2026-02-26},
	publisher = {arXiv},
	author = {Huang, Cheng and Tashi, Nyima and Gao, Fan and Liu, Yutong and Li, Jiahao and Tian, Hao and Jiang, Siyang and Tsering, Thupten and Ma-bao, Ban and Duojie, Renzeg and Luosang, Gadeng and Dongrub, Rinchen and Tashi, Dorje and Zhang, Jin and Feng, Xiao and Wang, Hao and Tang, Jie and Tang, Guojie and Wang, Xiangxiang and Zhang, Jia and Lee, Tsengdar and Yu, Yongbin},
	year = {2025},
	note = {Version Number: 1},
	keywords = {Computation and Language (cs.CL), FOS: Computer and information sciences},
}

@misc{kyogokuBasicModernTibetan2024,
	title = {Basic {Modern} {Tibetan} {SpaCy} {Model}},
	copyright = {Creative Commons Attribution 4.0 International},
	url = {https://zenodo.org/doi/10.5281/zenodo.14494472},
	doi = {10.5281/ZENODO.14494472},
	abstract = {This release contains a basic Modern Tibetan language model for SpaCy (trained with POS tags but no NER supported at the moment). The language model was trained to enable Tibetan-language support in the integrated Leipzig Corpus Miner (iLCM) and is thus trained on iLCM-compatible Python (3.7) and SpaCy (3.2.6) versions.

The model was trained by Yuki Kyogoku (Leipzig University) for the Divergent Discourses project. Computations for this work were done (in part) using resources of the Leipzig University Computing Center.

Full Changelog: https://github.com/Divergent-Discourses/Tibetan\_SpaCy-Model/commits/0.1.2},
	language = {bo},
	urldate = {2026-02-26},
	publisher = {Zenodo},
	author = {Kyogoku, Yuki and Erhard, Franz Xaver and Barnett, Robert and Hill, Nathan},
	month = dec,
	year = {2024},
	note = {Version Number: 0.1.2},
}

@article{kinadeterAnalyzingComparingMapping2026,
	title = {Analyzing, {Comparing} and {Mapping} {Zen} {Buddhist} {Genealogies} – {Facilitating} {Research} in {Buddhist} {Studies} by {Creating} a {Database}},
	url = {https://www.connections.clio-online.net/article/id/fda-160223},
	doi = {10.60693/TV2D-AJ54},
	language = {en},
	urldate = {2026-03-20},
	publisher = {Connections (Clio-online)},
	author = {Kinadeter, Michael},
	year = {2026},
	note = {Medium: text/html,application/pdf},
	keywords = {Area Studies, FOS: History and archaeology},
}

@inproceedings{nehrdichOneModelAll2024a,
	address = {Miami, Florida, USA},
	title = {One {Model} is {All} {You} {Need}: {ByT5}-{Sanskrit}, a {Unified} {Model} for {Sanskrit} {NLP} {Tasks}},
	url = {https://aclanthology.org/2024.findings-emnlp.805/},
	doi = {10.18653/v1/2024.findings-emnlp.805},
	abstract = {Morphologically rich languages are notoriously challenging to process for downstream NLP applications. This paper presents a new pretrained language model, ByT5-Sanskrit, designed for NLP applications involving the morphologically rich language Sanskrit. We evaluate ByT5-Sanskrit on established Sanskrit word segmentation tasks, where it outperforms previous data-driven approaches by a considerable margin and matches the performance of the current best lexicon-based model. It is easier to deploy and more robust to data not covered by external linguistic resources. It also achieves new state-of-the-art results in Vedic Sanskrit dependency parsing and OCR post-correction tasks. Additionally, based on the Digital Corpus of Sanskrit, we introduce a novel multitask dataset for the joint training of Sanskrit word segmentation, lemmatization, and morphosyntactic tagging tasks. We fine-tune ByT5-Sanskrit on this dataset, creating a versatile multitask model for various downstream Sanskrit applications. We have used this model in Sanskrit linguistic annotation projects, in information retrieval setups, and as a preprocessing step in a Sanskrit machine translation pipeline. We also show that our approach yields new best scores for lemmatization and dependency parsing of other morphologically rich languages. We thus demonstrate that byte-level pretrained language models can achieve excellent performance for morphologically rich languages, outperforming tokenizer-based models and presenting an important vector of exploration when constructing NLP pipelines for such languages.},
	booktitle = {Findings of the {Association} for {Computational} {Linguistics}: {EMNLP} 2024},
	publisher = {Association for Computational Linguistics},
	author = {Nehrdich, Sebastian and Hellwig, Oliver and Keutzer, Kurt},
	editor = {Al-Onaizan, Yaser and Bansal, Mohit and Chen, Yun-Nung},
	month = nov,
	year = {2024},
	pages = {13742--13751},
	file = {PDF:/home/thomas/Zotero/storage/GYVNGMVJ/Nehrdich et al. - 2024 - One Model is All You Need ByT5-Sanskrit, a Unified Model for Sanskrit NLP Tasks.pdf:application/pdf},
}

@article{zhangTibetanMedicalNamed2025,
	title = {Tibetan {Medical} {Named} {Entity} {Recognition} {Based} on {Syllable}‐{Word}‐{Sentence} {Embedding} {Transformer}},
	volume = {10},
	issn = {2468-6557, 2468-2322},
	url = {https://ietresearch.onlinelibrary.wiley.com/doi/10.1049/cit2.70029},
	doi = {10.1049/cit2.70029},
	abstract = {ABSTRACT
            Tibetan medical named entity recognition (Tibetan MNER) involves extracting specific types of medical entities from unstructured Tibetan medical texts. Tibetan MNER provide important data support for the work related to Tibetan medicine. However, existing Tibetan MNER methods often struggle to comprehensively capture multi‐level semantic information, failing to sufficiently extract multi‐granularity features and effectively filter out irrelevant information, which ultimately impacts the accuracy of entity recognition. This paper proposes an improved embedding representation method called syllable–word–sentence embedding. By leveraging features at different granularities and using un‐scaled dot‐product attention to focus on key features for feature fusion, the syllable–word–sentence embedding is integrated into the transformer, enhancing the specificity and diversity of feature representations. The model leverages multi‐level and multi‐granularity semantic information, thereby improving the performance of Tibetan MNER. We evaluate our proposed model on datasets from various domains. The results indicate that the model effectively identified three types of entities in the Tibetan news dataset we constructed, achieving an F1 score of 93.59\%, which represents an improvement of 1.24\% compared to the vanilla FLAT. Additionally, results from the Tibetan medical dataset we developed show that it is effective in identifying five kinds of medical entities, with an F1 score of 71.39\%, which is a 1.34\% improvement over the vanilla FLAT.},
	language = {en},
	number = {4},
	urldate = {2026-04-21},
	journal = {CAAI Transactions on Intelligence Technology},
	author = {Zhang, Jin and Zhang, Ziyue and Yeshi, Lobsang and Tashi, Dorje and Wang, Xiangshi and Cai, Yuqing and Yu, Yongbin and Wang, Xiangxiang and Tashi, Nyima and Luosang, Gadeng},
	month = aug,
	year = {2025},
	pages = {1148--1158},
	file = {PDF:/home/thomas/Zotero/storage/L4ITVNZI/Zhang et al. - 2025 - Tibetan Medical Named Entity Recognition Based on Syllable‐Word‐Sentence Embedding Transformer.pdf:application/pdf},
}

@inproceedings{liuActiveLearningTibetan2018,
	title = {Active {Learning} for {Tibetan} {Named} {Entity} {Recognition} based on {CRF}},
	booktitle = {Proceedings of the {LREC} 2018 {Workshop} {MLP}–{MoMent}},
	author = {Liu, Fei-Fei and Wang, Zhi-Juan},
	editor = {Du, Jinhua and Arcan, Mihael},
	year = {2018},
	pages = {18--45},
	file = {PDF:/home/thomas/Zotero/storage/2MM3UGXH/Liu und Wang - 2018 - Active Learning for Tibetan Named Entity Recognition based on CRF.pdf:application/pdf},
}

@article{luoNorbuKetakaAutoCorrecting2024,
	title = {Norbu {Ketaka}: {Auto}-{Correcting} {BDRC}'s {E}-{Text} {Corpora} {Using} {Natural} {Language} {Processing} and {Computer} {Vision} {Methods}},
	number = {72},
	journal = {Revue d'Etudes Tibétaines},
	author = {Luo, Queenie and van der Kuijp, Leonard W. J.},
	month = jul,
	year = {2024},
	pages = {26--42},
	file = {PDF:/home/thomas/Zotero/storage/3VWNZTAT/Luo und van der Kuijp - 2024 - Norbu Ketaka Auto-Correcting BDRC's E-Text Corpora Using Natural Language Processing and Computer V.pdf:application/pdf},
}

@article{meelenBreakthroughsTibetanNLP2024,
	title = {Breakthroughs in {Tibetan} {NLP} \& {Digital} {Humanities}},
	number = {72},
	journal = {Revue d'Etudes Tibétaines},
	author = {Meelen, Marieke and Nehrdich, Sebastian and Keutzer, Kurt},
	month = jul,
	year = {2024},
	pages = {5--25},
	file = {PDF:/home/thomas/Zotero/storage/3NYB4WR6/Meelen et al. - 2024 - Breakthroughs in Tibetan NLP & Digital Humanities.pdf:application/pdf},
}

@article{griffithsHandwrittenTextRecognition2024,
	title = {Handwritten {Text} {Recognition} ({HTR}) for {Tibetan} {Manuscripts} in {Cursive} {Script}},
	number = {72},
	journal = {Revue d'Etudes Tibétaines},
	author = {Griffiths, Rachael},
	month = jul,
	year = {2024},
	pages = {43--51},
	file = {PDF:/home/thomas/Zotero/storage/3IZ2TNRE/Griffiths - 2024 - Handwritten Text Recognition (HTR) for Tibetan Manuscripts in Cursive Script.pdf:application/pdf},
}

@article{faggionatoUniversalDependencyTreebank2024,
	title = {A {Universal} {Dependency} {Treebank} for {Classical} {Tibetan}},
	number = {72},
	journal = {Revue d'Etudes Tibétaines},
	author = {Faggionato, Christian},
	month = jul,
	year = {2024},
	pages = {52--69},
	file = {PDF:/home/thomas/Zotero/storage/K5BMZ8F2/Faggionato - 2024 - A Universal Dependency Treebank for Classical Tibetan.pdf:application/pdf},
}

@article{schmidtNLPReadabilityGraded2024,
	title = {{NLP} for {Readability}, {Graded} {Literature}, \& {Materials} {Development} in {Tibetan}},
	number = {72},
	journal = {Revue d'Etudes Tibétaines},
	author = {Schmidt, Dirk},
	month = jul,
	year = {2024},
	pages = {70--85},
	file = {PDF:/home/thomas/Zotero/storage/UA3F98MY/Schmidt - 2024 - NLP for Readability, Graded Literature, & Materials Development in Tibetan.pdf:application/pdf},
}

@article{nehrdich_observations_2023,
	title = {Observations on the {Intertextuality} of {Selected} {Abhidharma} {Texts} {Preserved} in {Chinese} {Translation}},
	volume = {14},
	issn = {2077-1444},
	url = {https://www.mdpi.com/2077-1444/14/7/911},
	doi = {10.3390/rel14070911},
	abstract = {Textual reuse is a fundamental characteristic of traditional Buddhist literature preserved in various languages. Given the sheer volume of preserved Buddhist literature and the often-unmarked instances of textual reuse, the thorough analysis and evaluation of this material without computational assistance are virtually impossible. This study investigates the application of computer-aided methods for detecting approximately similar passages within Xuanzang’s translation corpus and a selection of Abhidharma treatises preserved in Chinese translation. It presents visualizations of the generated network graphs and conducts a detailed examination of patterns of textual reuse among selected works within the Abhidharma tradition. This study demonstrates that the general picture of textual reuse within Xuanzang’s translation corpus and the selected Abhidharma texts, based on computational analysis, aligns well with established scholarship. Thus, it provides a robust foundation for conducting more detailed studies on individual text sets. The methods employed in this study to create and analyze citation network graphs can also be applied to other texts preserved in Chinese and, with some modifications, to texts in other languages.},
	language = {en},
	number = {7},
	urldate = {2026-05-21},
	journal = {Religions},
	author = {Nehrdich, Sebastian},
	month = jul,
	year = {2023},
	pages = {911},
	file = {PDF:/home/thomas/Zotero/storage/3VX9XS38/Nehrdich - 2023 - Observations on the Intertextuality of Selected Abhidharma Texts Preserved in Chinese Translation.pdf:application/pdf},
}

@article{bingenheimer_stylometric_2017,
	title = {Stylometric {Analysis} of {Chinese} {Buddhist} texts - {Do} different {Chinese} translations of the {Gaṇḍavyūha} reflect stylistic features that are typical for their age?},
	volume = {2},
	issn = {2188-7276},
	url = {https://www.jstage.jst.go.jp/article/jjadh/2/1/2_1/_article},
	doi = {10.17928/jjadh.2.1_1},
	language = {en},
	number = {1},
	urldate = {2026-05-21},
	journal = {Journal of the Japanese Association for Digital Humanities},
	author = {Bingenheimer, Marcus and Hung, Jen-Jou and Hsieh, Cheng-en},
	year = {2017},
	pages = {1--30},
	file = {Volltext:/home/thomas/Zotero/storage/N9FHRFAA/Bingenheimer et al. - 2017 - Stylometric Analysis of Chinese Buddhist texts - Do different Chinese translations of the Gaṇḍavyūha.pdf:application/pdf},
}

@misc{nehrdich_mitra_2026,
	title = {{MITRA}: {A} {Large}-{Scale} {Parallel} {Corpus} and {Multilingual} {Pretrained} {Language} {Model} for {Machine} {Translation} and {Semantic} {Retrieval} for {Pāli}, {Sanskrit}, {Buddhist} {Chinese}, and {Tibetan}},
	shorttitle = {{MITRA}},
	url = {http://arxiv.org/abs/2601.06400},
	doi = {10.48550/arXiv.2601.06400},
	abstract = {Ancient Buddhist literature features frequent, yet often unannotated, textual parallels spread across diverse languages: Sanskrit, Pāli, Buddhist Chinese, Tibetan, and more. The scale of this material makes manual examination prohibitive. We present the MITRA framework, which consists of a novel pipeline for multilingual parallel passage mining, MITRA-parallel, a large-scale corpus of 1.74 million parallel sentence pairs between Sanskrit, Chinese, and Tibetan, and the development of the domain-specific pretrained language model Gemma 2 MITRA. We present Gemma 2 MITRA-MT, a version of this base model fine-tuned on machine translation tasks, reaching state-of-the-art performance for machine translation of these languages into English and outperforming even much larger open-source models. We also present Gemma 2 MITRA-E, a semantic embedding model that shows state-of-the-art performance on a novel, detailed semantic embedding benchmark. We make the parallel dataset, model weights, and semantic similarity benchmark openly available to aid both NLP research and philological studies in Buddhist and classical Asian literature.},
	urldate = {2026-08-18},
	publisher = {arXiv},
	author = {Nehrdich, Sebastian and Keutzer, Kurt},
	month = jan,
	year = {2026},
	note = {arXiv:2601.06400 [cs.CL]},
	keywords = {Computer Science - Computation and Language},
	file = {Preprint PDF:/home/thomas/Zotero/storage/FS9SJYL7/Nehrdich und Keutzer - 2026 - MITRA A Large-Scale Parallel Corpus and Multilingual Pretrained Language Model for Machine Translat.pdf:application/pdf},
}