Merge pull request #33 from YerevaNN/Hrant-Khachatrian-patch-2

Update README.md

README.md

@@ -1,18 +1,22 @@
 # Chemlactica / Chemma: Large Language Models for Small Molecules
 
 TL;DR
-* A family of models that understand small organic molecules written in SMILES, their basic properties, and similarities between molecules.
-* [**Chemlactica-125M** 🤗](https://huggingface.co/yerevann/chemlactica-125m) and [**-1.3B** 🤗](https://huggingface.co/yerevann/chemlactica-1.3b) trained on top of Meta's [Galactica models](https://huggingface.co/facebook/galactica-1.3b).
+* A family of models that "understand" small organic molecules (SMILES), their basic properties (molecular weight, QED, SAS, TPSA, CLogP, ...), and similarities between molecules (Tanimoto over ECFC4).
+* [**Chemlactica-125M** 🤗](https://huggingface.co/yerevann/chemlactica-125m) and [**-1.3B** 🤗](https://huggingface.co/yerevann/chemlactica-1.3b) are trained on top of Meta's [Galactica models](https://huggingface.co/facebook/galactica-1.3b).
 * [**Chemma-2B** 🤗](https://huggingface.co/yerevann/chemma-2b) is built on top of Google's [Gemma-2B](https://huggingface.co/google/gemma-2b).
-* All models are trained on **40B** tokens covering 100M+ molecules from PubChem. [The dataset is also available at 🤗](https://huggingface.co/datasets/yerevann/PubChemForLM).
+* All models are trained on **40B** tokens covering 100M+ molecules from PubChem. [Check the corpus at 🤗](https://huggingface.co/datasets/yerevann/PubChemForLM).
 * A prompt like `</s>[SAS]2.25[/SAS][SIMILAR]0.62 CC(=O)OC1=CC=CC=C1C(=O)O[/SIMILAR][START_SMILES]` will generate a molecule that has ~2.25 SAS score and has ~0.62 similarity score to the given molecule.
-* The models can be easily tuned to perform property prediction (~0.3 RMSE on FreeSolv from MoleculeNet).
+* The models can be easily tuned to perform property prediction (~0.3 RMSE on [FreeSolv](https://paperswithcode.com/sota/molecular-property-prediction-on-freesolv) from MoleculeNet).
 * The models wrapped into a **genetic-like optimization algorithm** beat all **molecular optimization** benchmarks we tried.
-  * [**Practical Molecular Optimization**](https://arxiv.org/abs/2206.12411): **17.5** vs 16.2 (previous SOTA: [Genetic-guided GFlowNets](https://arxiv.org/abs/2402.05961)).
-  * Optimization for **docking** with AutoDock Vina: 3-4x less oracle calls for generating 100 _good_ molecules than previous SOTA.
-  * QED optimization from the [RetMol paper](https://arxiv.org/abs/2208.11126): **99%** success rate with 10K oracle calls with Chemlactica-125M (vs. 96% with 50K calls).
-* All details in the paper [Small Molecule Optimization with Large Language Models](https://yerevann.com/papers/small-molecule-optimization-with-large-language-models).
-
+  * [**Practical Molecular Optimization**](https://arxiv.org/abs/2206.12411)
+    * **17.5** vs 16.2 (previous SOTA: [Genetic-guided GFlowNets](https://arxiv.org/abs/2402.05961)).
+  * Optimization for **docking** with AutoDock Vina
+    * 3-4x fewer oracle calls for generating 100 _good_ molecules than previous SOTA ([Beam Enumeration](https://arxiv.org/abs/2309.13957)).
+  * QED optimization from the [RetMol paper](https://arxiv.org/abs/2208.11126)
+    * **99%** success rate with 10K oracle calls with Chemlactica-125M (vs. 96% with 50K calls of the original paper).
+* Read the details in the paper [_Small Molecule Optimization with Large Language Models_](https://yerevann.com/papers/small-molecule-optimization-with-large-language-models.pdf).
+
+We are looking forward to the community utilizing these models for solving various problems in molecular design.
 
 ## Table of contents
 - [Description](#Description)