Add citation: Findings of EMNLP 2026

README: drop checkpoint epoch/step numbers

Drop the last internal run id from a config comment

Use role-based names instead of internal experiment run names

Replace author-specific absolute paths in configs with neutral placeholders

Replace R16 tokenizer with final LS+Emilia checkpoint (epoch_22)

Upload README.md with huggingface_hub

Upload README.md with huggingface_hub

rename word_asr -> streaming_asr

rename streaming_asr -> alignment_model

Upload word_asr/chunk_streaming_word_fastemit.yaml with huggingface_hub

Upload word_asr/tokenizer.ckpt with huggingface_hub

Upload folder using huggingface_hub

Upload boundary_classifier/boundary_classifier.yaml with huggingface_hub

Upload boundary_classifier/best_precision_model.pt with huggingface_hub

Upload boundary_classifier/best_model.pt with huggingface_hub

Upload alignment.yaml with huggingface_hub

Upload folder using huggingface_hub

Upload rvq_teacher/epoch_16.pt with huggingface_hub

Upload README.md with huggingface_hub

initial commit