All pull requests

Pull requests

• [LLM] Show the other readings of a word on the peek card

  #42·menny opened Nov 8, 2026Member·17/17·

  ·

• [LLM] Choose the primary reading of common ambiguous spellings by hand

  #41·menny opened Nov 8, 2026Member·16/17·

  ·

• [LLM] Keep every reading of a spelling in the dictionary

  #40·menny opened Nov 8, 2026Member·15/17·

  ·

• [LLM] Add four micro-stories for levels 1 and 2

  #39·menny opened Nov 8, 2026Member·14/17·

  ·

• [LLM] Add 13 false cognates to the curated seed

  #38·menny opened Nov 8, 2026Member·13/17·

  ·

• [LLM] Keep proper nouns from hiding common words and fix two seed lemmas

  #37·menny opened Nov 8, 2026Member·12/17·

  ·

• [LLM] Add the extracted SUBTLEX, Kaikki and Tatoeba subsets and build seed.db from them

  #36·menny opened Nov 8, 2026Member·11/17·

  ·

• [LLM] Generate seed.db with Bazel instead of checking it in

  #35·menny opened Nov 8, 2026Member·10/17·

  ·

• [LLM] Validate the curated seed JSON files and fix a disallowed term

  #34·menny opened Nov 8, 2026Member·9/17·

  ·

• [LLM] Group Tatoeba sentences into leveled reading lessons

  #33·menny opened Nov 8, 2026Member·8/17·

  ·

• [LLM] Audit cognate detection, false-cognate expansion and sentence scoring

  #32·menny opened Nov 8, 2026Member·7/17·

  ·

• [LLM] Make KaikkiProcessor read the lemma subset and its aliases

  #31·menny opened Nov 8, 2026Member·6/17·

  ·

• [LLM] Keep the best dictionary row per surface form and fix SUBTLEX header handling

  #30·menny opened Nov 8, 2026Member·5/17·

  ·

• [LLM] Add Tatoeba candidate extractor

  #29·menny opened Nov 8, 2026Member·4/17·

  ·

• [LLM] Add Kaikki lemma-subset extractor with aliases

  #28·menny opened Nov 8, 2026Member·3/17·

  ·

• [LLM] Add SUBTLEX-ESP extractor, preprocessor README and NOTICE

  #26·menny opened Nov 8, 2026Member·2/17·

  ·

• [LLM] Update TODO.md with the data and ingestion plan

  #25·menny opened Nov 8, 2026Member·1/17·

  ·