Repository navigation
[LLM] Implement Pipeline 2 on-device text enrichment engine - #9
Merged
Merged
Conversation
- Add ReaderEnricher for single-pass Unicode-aware token segmentation and interactive TokenSpan enrichment. - Add CliticDeconstructor implementing clitic pronoun surgery (accent restoration and compound pronoun breakdown). - Add NumberConverter for deterministic Spanish number conversion (0..999,999) with gender agreement and apocope. - Add EnrichmentModels defining TokenSpan, CliticBreakdown, DictionaryEntry, FalseCognateEntry, and WordLookupProvider. - Add comprehensive test suites for CliticDeconstructor, NumberConverter, and ReaderEnricher.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
Implements Pipeline 2 (On-Device Runtime Text Ingestion) and associated linguistic algorithms as specified in docs/INGESTION.md and docs/DATA.md:
BreakIteratorword segmentation. Produces interactiveTokenSpanobjects while preserving exact string offsets.diciéndomelo->diciendo+me+lo), restoring stress accents (á, é, í, ó, ú).un,veintiún).TokenSpan,CliticBreakdown,DictionaryEntry,FalseCognateEntry, and theWordLookupProviderinterface.CliticDeconstructorTest,NumberConverterTest,ReaderEnricherTest).Stacked on top of #7 (
pipeline-1-preprocessor).