Card: positioning alt text matches the trained-parameters axis

model card: positioning chart in cross-modal section

model card: model overview figure

Note the safetensors/param-count meaning

Add badges under the banner

Display banner image on the model card

Add tagline banner to model card

MAEB: exact FSDKaggle2019 overlap numbers (608/4481 verified by Freesound id)

MAEB: FSD2019Kaggle disclosure (anonymized filenames, disjointness unverifiable)

Add banner to model card

Restructure card: architecture up top, unified-model comparison leads evaluation

MAEB(beta) results summary

Add Gemini Embedding 2 baseline to cross-modal comparison

Add LanguageBind baseline to cross-modal comparison

v0.3 model card with three-version comparison

Model card: architecture diagram; fix bank size to 484K (v0.2)

Document version tags

v0.2 retrieval examples in model card

Bold best-per-column only in comparison tables

Bold v0.2 numbers in comparison tables

v0.2 model card with v0.1 comparison

Add VGGSound attribution to gallery

Embed retrieval gallery in model card

Add full cross-modal metrics and qualitative retrieval examples

Upload README.md with huggingface_hub