docs: refresh attribution for full model catalog

README.md

@@ -1,8 +1,10 @@
-# sparse-embeddings
+# Sparse Embeddings
 
-Self-hosted ONNX models for [xberg](https://github.com/xberg-io/xberg).
-Mirrored from upstream permissive (Apache-2.0/MIT) sources; weights unmodified.
+Self-hosted mirror of permissively-licensed retrieval models used by [xberg](https://github.com/xberg-io/xberg). Weights are unmodified; every file is sha256-pinned in the xberg source tree (`crates/xberg/src/sparse_embeddings/presets.sha256sum`) and verified at download.
 
-## Contents & attribution
+All models are Apache-2.0 or MIT licensed. Attribution to each upstream author below.
 
-- **`splade/`** — prithivida/Splade_PP_en_v1 (Apache-2.0). License: Apache-2.0.
+| Model | Upstream | License | Notes |
+|---|---|---|---|
+| `splade` | [prithivida/Splade_PP_en_v1](https://huggingface.co/prithivida/Splade_PP_en_v1) | Apache-2.0 | SPLADE learned sparse |
+| `opensearch-v3-distill` | [opensearch-project/opensearch-neural-sparse-encoding-doc-v3-distill](https://huggingface.co/opensearch-project/opensearch-neural-sparse-encoding-doc-v3-distill) | Apache-2.0 | SPLADE sparse (default) |