docs: add Bekko blog and paper links

README.md

@@ -36,7 +36,7 @@ For a smaller, faster model, see [bekko-embedding-v1-a8m](https://huggingface.co
 You can also try bekko right in your browser: the [bekko-embedding-web](https://huggingface.co/spaces/hotchpotch/bekko-embedding-web) demo runs the model fully client-side with Transformers.js — no server involved.
 
 > [!NOTE]
-> For technical details, including the model architecture and training recipe, see the [Bekko Embedding paper](https://huggingface.co/papers/2607.25180).
+> For a guided overview of the models, training recipe, and results, read [Bekko Embedding: how small can a multilingual retrieval model be?](https://huggingface.co/blog/hotchpotch/bekko-embedding).
 
 ## Highlights
 
@@ -627,6 +627,10 @@ How much quality do you trade for a smaller index? For `bekko-embedding-v1-a25m`
 
 The name pairs the protective spirit of the red ox with the quiet beauty of this classic amber tone.
 
+## Paper
+
+For full technical details, see [Bekko Embedding: Parameter-Efficient Multilingual Retrieval with Ultra-Compact Encoders](https://huggingface.co/papers/2607.25180).
+
 ## Citation
 
 If you use bekko-embedding in your work, please cite: