Update README.md

README.md

@@ -4,10 +4,10 @@ language:
 license: cc-by-nc-4.0
 ---
 ## Introduction
-We introduce MMEmbed, an extension of NV-Embed-v1 with multimodal retrieval capability. 
-MMEmbed achieves state-of-the-art results in [UniIR benchmark](https://huggingface.co/TIGER-Lab/UniIR) with 52.7 averaged score compared to 48.9 (the best results in [UnIR benchmark paper](https://eccv.ecva.net/virtual/2024/poster/863)). 
-Notably, MMEmbed improves NV-Embed-v1 text retrieval accuracy, from 59.36 to 60.3 on 15 retrieval tasks within Massive Text Embedding Benchmark ([MTEB benchmark](https://arxiv.org/abs/2210.07316)).
-MMEmbed presents several new training strategies, including modality-aware hard negative mining to improve multimodal retrieval accuracy in UniIR, and demonstrating a continual text-to-text fine-tuning method to further enhance the accuracy of text-to-text retrieval while maintaining mulitmodal retrieval accuracy.
+We introduce MM-Embed, an extension of NV-Embed-v1 with multimodal retrieval capability. 
+MM-Embed achieves state-of-the-art results in [UniIR benchmark](https://huggingface.co/TIGER-Lab/UniIR) with 52.7 averaged score compared to 48.9 (the best results in [UnIR benchmark paper](https://eccv.ecva.net/virtual/2024/poster/863)). 
+Notably, MM-Embed improves NV-Embed-v1 text retrieval accuracy, from 59.36 to 60.3 on 15 retrieval tasks within Massive Text Embedding Benchmark ([MTEB benchmark](https://arxiv.org/abs/2210.07316)).
+MM-Embed presents several new training strategies, including modality-aware hard negative mining to improve multimodal retrieval accuracy in UniIR, and demonstrating a continual text-to-text fine-tuning method to further enhance the accuracy of text-to-text retrieval while maintaining mulitmodal retrieval accuracy.
 
 <!-- For more technical details, refer to our paper: [NV-Embed: Improved Techniques for Training LLMs as Generalist Embedding Models](https://arxiv.org/pdf/2405.17428). -->