MITRA-lexicon: 2.18M-entry Sanskrit-English-Tibetan-Chinese lexica built with LLM alignment
SebastianNehrd2 · x · 2026-10-20
Researchers including Sebastian Nehrdich released the MITRA-lexicon preprint, a family of corpus-grounded lexica aligning Sanskrit with English, Tibetan, and Chinese.
- Scale: 2.18M headword entries and 10.4M attestations across six directional lexica, each linked to its sentence pair with character offsets
- Sources: the Mitrasamgraha corpus (Sanskrit-English) and the MITRA-parallel Buddhist corpus (Sanskrit-Tibetan/Chinese)
- Method: a multi-step pipeline combining grammatical annotation models, fine-tuned LLMs, and a teacher-student setup
- Released with a web interface, the lexica data, and the fine-tuned span annotation model
Related event: MITRA Lexicon Released with Web App for Buddhist Texts(2 posts)→
More from Research
- New paper: let models reward hack in RL, then disable the parameters at deployment — nptacek · 2026-10-20
- Victor Taelin: smart enough compilers will understand and swap your algorithms — bronzeagepapi · 2026-10-20
- Galbot proposes observational scaling law for physical recursive self-improvement in VLM agents — Galbot · 2026-10-20
- ActionRoPE encodes action as world coordinates for spatially consistent game world models — Zhiyang Deng · 2026-10-20
- UniAVC unifies audio-video generation via progressive training and asymmetric noise scheduling — Kien T. Pham · 2026-10-20
- AHE paper auto-evolves coding-agent harnesses with decoupled files, lifting Terminal-Bench 2 to 77% — morgymcg · 2026-10-20