weemed/IlhaEmbed

feat(brand): add 17th C. Ilha Formosa horizontal nautical line-art logo to header

feat(brand): add 17th C. Ilha Formosa horizontal nautical line-art logo asset

fix(style): remove block img from headings to ensure clean typography and proper heading alignment

style(readme): replace informal emojis with clean institutional vector SVG icons

feat(assets): add vector SVG icon scale.svg

feat(assets): add vector SVG icon shield.svg

feat(assets): add vector SVG icon terminal.svg

feat(assets): add vector SVG icon clipboard.svg

feat(assets): add vector SVG icon flask.svg

feat(assets): add vector SVG icon chart.svg

feat(assets): add vector SVG icon sparkles.svg

docs: add official MTEB benchmark comparison and 20-note clinical evaluation report

docs: fix CJK markdown bold rendering and optimize community tags with interactive widget

docs: restore canonical English-first Chinese-second README and replace model_int8.onnx with 37MB compact edge model

clean: remove internal research artifacts and streamline model card

docs: restore clean 311M flagship benchmarks and isolate 97M edge retraining table

Release IlhaEmbed-97M Edge v4: Multi-Head Attention Pooling & RxNorm 85.99%

Release IlhaEmbed-97M Edge: 100% (44/44) FHIR prototype routing & INT8 ONNX

docs: update model card benchmark to 100% FHIR routing for Edge 97M

docs: sync model card from WeeMed/ilhaembed

docs: sync model card from WeeMed/ilhaembed

Correct v2.1.0 status: automatic classification is not qualified

Release IlhaEmbed v2.1.0 with corrected category regression evaluation

docs: remove internal legacy CODER-TW references from SOURCES.md

docs: update README.md Model Card with v2.0 clinical specs and benchmark tables

release: update IlhaEmbed with expanded clinical lab & procedural contrastive tuning

release: update IlhaEmbed with expanded clinical lab & procedural contrastive tuning

release: update IlhaEmbed with expanded clinical lab & procedural contrastive tuning

release: update IlhaEmbed with expanded clinical lab & procedural contrastive tuning

release: update IlhaEmbed with expanded clinical lab & procedural contrastive tuning

Release corrected source-isolated IlhaEmbed v2

docs: update Hugging Face main README model card for IlhaEmbed v2.0

docs: update Hugging Face main README model card for IlhaEmbed v2.0

feat: update model_int8.onnx (IlhaEmbed v2 NAER retrained INT8 ONNX)

feat: upload IlhaEmbed v2 model weights and tokenizer (NAER 13-set retrained)

docs: add Context-Conditioned usage and MODA Data Provenance to Traditional Chinese section

docs: fix strikethrough markdown formatting caused by unescaped tildes

docs: update Model Card with Context-Conditioned usage and MODA provenance

docs: sync model card from WeeMed/ilhaembed

docs: definitive eval numbers (semantic macro 0.86); reframe taigi as a transliteration diagnostic (not a self-match bug), excluded from the macro with reason

docs: honest metrics correction (apposition 0.87 not 0.90; disclose taigi self-match eval bug and PTT noise; show full 0.52 all-register macro)

rewrite: natural Taiwan wording, understated tone, correct spacing

add top language nav (English/繁體中文 jump)

model card: natural voice, no em-dashes, real numbers

IlhaEmbed-97M (vocab-pruned) initial release

initial commit