Updated README.md

README.md

@@ -25,7 +25,7 @@ This repository provides PyTorch source code associated with our publication, "S
 
 **Paper:** [OpenReview Link](https://openreview.net/pdf?id=0uWNuJ1xtz)
 
-**HuggingFace:** [HuggingFace Link](https://huggingface.co/ibm/materials.str-bamba)
+**GitHub:** [GitHub Link](https://github.com/IBM/materials/tree/main/models/str_bamba)
 
 For more information contact: vshirasuna@ibm.com or evital@br.ibm.com.
 
@@ -33,7 +33,7 @@ For more information contact: vshirasuna@ibm.com or evital@br.ibm.com.
 
 ## Introduction
 
-We present a large encoder-decoder chemical foundation model based on the IBM Bamba architecture, a hybrid of Transformers and Mamba-2 layers, designed to support multi-representational molecular string inputs. The model is pre-trained in a BERT-style on 588 million samples, resulting in a corpus of approximately 29 billion molecular tokens. These models serve as a foundation for language chemical research in supporting different complex tasks, including molecular properties prediction, classification, and molecular translation. **Additionally, the STR-Bamba architecture allows for the aggregation of multiple representations in a single text input, as it does not contain any token length limitation, except for hardware limitations.** Our experiments across multiple benchmark datasets demonstrate state-of-the-art performance for various tasks. Model weights are available at: [HuggingFace Link](https://huggingface.co/ibm/materials.str-bamba).
+We present a large encoder-decoder chemical foundation model based on the IBM Bamba architecture, a hybrid of Transformers and Mamba-2 layers, designed to support multi-representational molecular string inputs. The model is pre-trained in a BERT-style on 588 million samples, resulting in a corpus of approximately 29 billion molecular tokens. These models serve as a foundation for language chemical research in supporting different complex tasks, including molecular properties prediction, classification, and molecular translation. **Additionally, the STR-Bamba architecture allows for the aggregation of multiple representations in a single text input, as it does not contain any token length limitation, except for hardware limitations.** Our experiments across multiple benchmark datasets demonstrate state-of-the-art performance for various tasks. Model weights are available at: [GitHub Link](https://github.com/IBM/materials/tree/main/models/str_bamba).
 
 The STR-Bamba model supports the following **molecular representations**:
 - SMILES
@@ -60,7 +60,7 @@ The STR-Bamba model supports the following **molecular representations**:
 
 ### Pretrained Models and Training Logs
 
-We provide checkpoints of the STR-Bamba model pre-trained on a dataset of ~118M small molecules, ~2M polymer structures, and 258 formulations. The pre-trained model shows competitive performance on classification and regression benchmarks across small and polymer molecules, and electrolyte formulations. For model weights: [HuggingFace Link](https://huggingface.co/ibm/materials.str-bamba)
+We provide checkpoints of the STR-Bamba model pre-trained on a dataset of ~118M small molecules, ~2M polymer structures, and 258 formulations. The pre-trained model shows competitive performance on classification and regression benchmarks across small and polymer molecules, and electrolyte formulations. For model weights: [GitHub Link](https://github.com/IBM/materials/tree/main/models/str_bamba)
 
 Add the STR-Bamba `pre-trained weights.pt` to the `inference/` or `finetune/` directory according to your needs. The directory structure should look like the following: