jh-y
/

dllm-vsr

visual-speech-recognition

discrete-diffusion

diffusion-language-model

Model card Files Files and versions

dllm-vsr / usr2 /len_pred

Commit History

Update model card (EMNLP 2026 camera-ready) and add length-predictor outputs (K_pred±5 pools for LRS3 test / WildVSR)

7a9cd42
verified

jh-y commited on Sep 2

Upload folder using huggingface_hub

78f8a52
verified

jh-y commited on May 27