Sebastian Nehrdich

Thrilled to share that I will be joining Tohoku University in Sendai, Japan, as tenure track assistant professor from autumn this year. Happy to work on AI for Sanskrit, Chinese, Tibetan, Pali, and Japanese, and to educate new generation of students who can work with these tools.

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 have recently changed our etext basis for Kangyur and Tengyur to the Esukhia Derge version, making folio-level linking to the ToUDA scanned Derge edition and to rKTs possible via the new "segment view" option in Mitra Deep Research & Explore!

@5chstereoAI you can always send us an email at dharmamitra.project@gmail.com about feature requests and feedback! We are happy to learn more and to explore how we can improve the platform.

😅 I just saw that you are Prof at Tohoku! I use your tools for research and they are very helpful! Do you have a link where I can submit feedback / feature request? For example, I think it would be very helpful to have a GIS type map showing how Dhamma travelled across time and geography.

Very modern technology meets very traditional scholarship! To the people who made those translations under painstaking effort int hose years, it must sound like absolute science fiction what we can do with them nowadays!

@5chstereoAI We are already working together with our archive unit here At Tohoku university and just added links to the new Derge database!

@SebastianNehrd2 Do you also partner with other universities? x.com/ag0vb/status/2…

서울대 교양채널 샤로잡다에서 힙불교 열풍, 불교는 종교인가 철학인가, 불교철학의 매력, 불교의 기본적 교리 등의 주제를 논하고 왔습니다. youtu.be/a6l44RlgcPU?si…

We just rolled out a massive update to our interface over the last days! One of the massive changes is that we replaced MITRA Search with MITRA Explore, which brings deep research functionality into the Search system with filter options and more 1/

there certainly is, the problem is more on the structural end, i.e. we don't usually have clean text-level parallels like in the case of the tibetan translation of the Ybh etc. so the matching becomes less precise and reliable, and Pali literature has a lot of repititions etc. which make automatize matching a bit more tricky! But I think we can do this in a future update.

@SebastianNehrd2 Amongst the Agamas, Nikayas and Vinaya, is there enough correspondence to do a partial equivalent including Pali?

Since I repeatedly got the request on whether there is a way to just browse the mitra parallel data without all the DharmaNexus matching etc. around it, I created a small html page where you can do exactly that! dharmamitra.github.io/mitra-parallel…

Dharmamitra is very concerned about the implications of machine translation software on the language abilities of the users! We therefore now introduce a new feature that will ask the users to manually translate a passage every five requests. 1/

Happening this weekend, online participation possible!

MITRA Explore is coming and it will be 🔥🔥

Coming very soon: MITRA Explore will enable to ask more open questions and get answers based on the powerful retrieval capabilities of Dharmamitra!

One recent observation by a colleague: “cracking complex Sanskrit sentences, understanding their grammar and translating them used to be a solitary joy. AI tools like dharmamitra make it less solitary, and less joyful.”

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 of highest quality.

A bit late to the party but this is a great paper, and I am very happy to see that ByT5-Sanskrit is indeed a versatile model that adepts well and outperforms much larger LLMs in task-specific settings for Sanskrit!

This is sooo avoidable! Like there is zero intellectual effort in looking up the papers and citing them properly, at least that much effort one can expect right!

I am reviewing for various computer science conferences these days and I had to reject roughly 40% of the papers I looked at so far purely on the fact that their bibliographies are filled with hallucinations.

2. Mitrasamgraha: A Comprehensive Classical Sanskrit Machine Translation Dataset arxiv.org/abs/2601.07314 A large dataset of parallel sentence pairs for Classical/Vedic Sanskrit to English, covering multiple domains and time spans. Useful for machine translation!

Two preprints of the Dharmamitra project: 1. MITRA: arxiv.org/abs/2601.06400 This paper describes our large multilingual parallel dataset release, the machine translation model, and our retrieval system. 1/