Update README.md

README.md

@@ -38,7 +38,7 @@ base_model: sentence-transformers/LaBSE
 
 ## What is NE-Embed?
 
-NE-Embed is the first text embedding model purpose-built for Northeast Indian languages. It enables **semantic search, retrieval, and RAG** across languages that general multilingual models fail on — languages like Garo, Meitei, Nyishi, and Pnar where raw LaBSE achieves under 15% retrieval accuracy.
+NE-Embed A multilingual text embedding model purpose-built for Northeast Indian languages. It enables **semantic search, retrieval, and RAG** across languages that general multilingual models fail on — languages like Garo, Meitei, Nyishi, and Pnar where raw LaBSE achieves under 15% retrieval accuracy.
 
 Trained on 201,738 balanced parallel pairs across 10 languages using bi-encoder fine-tuning with `MultipleNegativesRankingLoss`.
 
@@ -89,7 +89,7 @@ Evaluated on 500 samples per language. CLRI = Cross-Language Retrieval Interfere
 ```python
 from sentence_transformers import SentenceTransformer
 
-model = SentenceTransformer("MWirelabs/ne-embed-v1")
+model = SentenceTransformer("MWirelabs/ne-embed")
 
 sentences = [
     "Where is the nearest hospital?",                          # English
@@ -117,7 +117,17 @@ score = 0.7 * ne_embed_score + 0.3 * bm25_score
 - **Loss:** `MultipleNegativesRankingLoss`
 - **Data:** 201,738 English↔NE language parallel pairs, capped at 25k per language to prevent Assamese attractor bias
 - **Epochs:** 3 · **Batch size:** 64 · **Max seq length:** 128
-- **Hardware:** NVIDIA A40 · **Training time:** ~1.3 hours
+- **Hardware:** 1× NVIDIA A40 (48 GB) · **Training time:** ~1.3 hours
+
+---
+
+## Intended Uses
+
+- Semantic search
+- Dense retrieval
+- RAG
+- Cross-lingual retrieval
+- Clustering
 
 ---