For anybody interested in a tentative dating of the sanskrit literature included at Dharmamitra.org, based on linguistic features and human estimated priors, grouped by categories!
Sebastian Nehrdich
@snehrdich.bsky.social
CTO of the MITRA project @BAIR, UC Berkeley. Research in ancient Asian low resource languages, especially text reuse, machine translation, semantic similarity search. Buddhist studies MA, now PhD in computational linguistics @Duesseldorf university.
This deserves attention! A whole book written by a Japanese Studies expert on a topic of Korean history relies heavily on AI tools for accessing the sources and acknowledges it clearly in the preface. Is this the kind of future we want for humanities?
We are happy to announce that Dharmamitra now features a board of advisors. They will advise on the kind of data that Dharmamitra includes, on the functionality and design of our applications, and on making sure that we keep providing tools and utility that really matter for our core audience.
Dharmamitra has recently seen the integration of the fantastic Digital Dictionary of Buddhism for Chinese and the equally great dictionary for Tibetan by Christian Steinert into the “English (explained)” translation mode!
My talk at the Khyentse Foundation Goodman lecture series about the relationship between AI and philological Buddhist Studies, and how Dharmamitra fits into this new emerging landscape: youtu.be/Ezxg9TwRf5g
The Goodman Lecture 32 | Buddhist Philology and AI
YouTube video by Khyentse Foundation
youtu.be
Chatting with my colleague Zorg of the Dharmamitra team about our most recent platform update and the launch of DharmaNexus: youtu.be/cgF9NvnGXG8
Chat About Dharmamitra Updates in 2025 and DharmaNexus
YouTube video by Dharmamitra
youtu.be
dharmamitra.org just saw a huge set of updates the last days: - Addition of MITRA Search - Addition of MITRA Deep Research Translation - Launch of dharmanexus at dharmanexus.org, You can learn more here: dharmamitra.github.io/dharmamitra-...
Dharmamitra · a toolkit for the languages of the Dharma
A toolkit for the languages of the Dharma
dharmamitra.org
Dharmamitra got a significant update: We now feature fast OCR for Sanskrit, Tibetan, etc. powered by Gemini. You can upload images and PDFs. We also added an option to translate from files directly, instead of needing to go through OCR manually first!
I will be giving a presentation on the effectiveness of semantic similarity models for textual reuse detections in Buddhist source languages with focus on Buddhist Chinese in this great online workshop coming up at Bochum University! www.oaw.ruhr-uni-bochum.de/forschung/ha...
Happy to announce that I will be joining Tohoku University as tenure track assistant professor in autumn this year! My position will be at the intersection of Buddhist & Japanese Studies and machine learning / digital humanities.
I will be presenting the various MITRA tools and what exciting new features we will offer in 2025 at the IABS conference in Leipzig on Tuesday August 12 in this panel organized by Marcus Bingenheimer!
sites.google.com/view/dhws2025b I will be in Tokyo from the 12th to the 17th to talk about Dharmamitra and how we got to where we are now at Keio University.
DH国際ワークショップ
2025年6月13日(金曜)10:45-13:30 慶應義塾大学三田キャンパス東館G-lab 同時通訳付き/ハイブリッド開催 参加申込みフォーム(無料、要事前申込み) 共催 科研費特別推進研究「デジタル研究基盤としての令和大蔵経の編纂―次世代人文学の研究基盤構築モデルの提示(JP25H00001)」 人文学・社会科学のDX化に向けた研究開発推進事業・研究基盤ハブ
sites.google.com
The massive MITRA Sanskrit-Tibetan and Tibetan-Sanskrit dictionary with more than 4M entries based on 600,000 sentence pairs is ready for public download! github.com/dharmamitra/...
GitHub - dharmamitra/dharmamitra-stardict-dictionaries: Stardict Dictionaries based on Dharmamitra parallel data collection efforts
Stardict Dictionaries based on Dharmamitra parallel data collection efforts - GitHub - dharmamitra/dharmamitra-stardict-dictionaries: Stardict Dictionaries based on Dharmamitra parallel data colle...
github.com
Our comprehensive work on evaluation of machine translation quality for Buddhist Chinese, including comparision of different data augmentation statergies, is accepted at NLP4DH, proceedings will come soon: www.academia.edu/128624124/MI...
MITRA-zh-eval: Using a Buddhist Chinese Language Evaluation Dataset to Assess Machine Translation and Evaluation Metrics
With the advent of large language models, machine translation (MT) has become a widely used, but little understood, tool for accessing historical and multilingual texts. While models like GPT, Claude,...
academia.edu
I had a great time teaching a workshop on LLMs and AI technology for Asian studies at the AAS conference in Columbus, Ohio this year!
The author decides a priori that the Indus Valley seals represent a language and then decides that the language of the Indus seals is Sanskrit. Why? He can exclude all agglutinative languages (which include for them Sumerian and Dravidian etc.), because: 3/
A Summer school in Philosophical #Sanskrit: The 4th edition of the Vienna Summer School on the Cultural and Intellectual History of Asia will take place in Vienna, Austria, from July 7 to July 12 2025. (Fees are ridiculously low: 100 E w/o accommodation or 400 w accommodation & breakfast) 1/
IKGA
oeaw.ac.at
The world is trim enough these days. Here is the view of downtown San Francisco and the bay bridge taken from Kensington.
The recordings of the conference "AI and the Future of Buddhist Studies" held at UC Berkeley are finally online: www.youtube.com/watch?v=4JKd...
Day 1: AI and the Future of Buddhist Studies
YouTube video by Numata Center for Buddhist Studies, UC Berkeley
youtube.com
Not many people will like this but with the advent of AI/LLM technology, we are facing a future where philology and translation as a main profession for Buddhist Studies will quickly disappear or fade to the point of irrelevance over the next 10-20 years. 1/
Gemini 2 is perhaps the best available AI model out there that few people talk about. In our tests on translating Classical/Buddhist Chinese into English, Gemini 2 Flash + RAG augmentation is competitive to Claude 3.5 Sonnet while being more than a hundred times cheaper. Google is 🔥🔥
Living in the USA during the beginning of Trumps second term makes it much easier for me to understand how the McCarthy era could happen in a liberal democratic system
The Dharmamitra chrome extension also got a significant update. In addition to a better underlying model, It now supports multiple output languages. Just right-click on the dharmamitra icon, open "options" and then set it to the language you prefer. Enjoy!
Dharmamitra just got a whole lot better at just about everything, especially when it comes to translating into or out of non-English languages! It’s still not perfect but it now again beats claude etc. by a margin. Bigger update will come with new functionality in the coming days, stay tuned!
Chinese New Year's gift: MITRA will very soon significantly increase its capabilities across the board, including translation into modern Chinese. Its going to be 🔥🔥
Japan has always been very kind to me. In 2018, Tsukuba university invited me for a young scholars workshop, that was my first visit. From 2018-2020 I spent excellent years in Kyoto. Now I could come back for the 100 year celebration of the 大正大蔵経/30 years of SAT. ありがとう!
My talk on dharmamitra at the Heidelberg/Otani online lecture series is available on YouTube for those of you who are interested!
My talk on dharmamitra, what it can do now and what it will be able to do in the future this week! Afaik it is going to be public, via the QR code in the announcement you should be able to join.