There’s some Sanskrit texts that don’t have English translations that I’d really like to read so I was going to use an LLM to create translations. Does anyone know how well LLMs handle Sanskrit or have any suggestions? Are any particular models better than others? Especially because I know that ancient texts sometimes use different dialects of Sanskrit and have different challenges.
The one I’m mostly interested in translating is called the Moksopaya, here’s the Sanskrit on GRETIL: https://gretil.sub.uni-goettingen.de/gretil/1_sanskr/6_sastr...
No, LLMs do not handle low resource / low corpus languages, including Sanskrit well.
Yesterday in HN "what are you working on", dr_dshiv presented SourceLibrary [1] which has tons of machine translation of beautiful texts.
Here's my "Let Me SourceLibrary That For You" using their Librarian search which presents the "The Yoga Vāsiṣṭha (The Expanded Mokṣopāya)" [2]
ETA: My apologies, I went deeper and the links therein are scans/ocr of English translations of that text. The library does have other Sanskrit translations, but not of that?
[1] https://news.ycombinator.com/item?id=48889335
[2] https://sourcelibrary.org/librarian?thread=6a56c578a50845441...