Update README.md

README.md

@@ -10,7 +10,7 @@ By leveraging MoE layers, AIDO.Protein efficiently scales to 16 billion paramete
 
 ## Model Architecture Details
 ADIO.Protein is a transformer encoder-only architecture with the dense MLP layer in each transformer block replaced by a sparse MoE layer. It uses single amino acid tokenization and is optimized using a masked languange modeling (MLM) training objective. For each token, 2 experts will be selectively activated by the top-2 rounting mechiansim.
-<center><img src="Proteion-MOE architecture.png" alt="An Overview of AIDO.Protein" style="width:70%; height:auto;" /></center>
+<center><img src="proteinmoe_architecture.png" alt="An Overview of AIDO.Protein" style="width:70%; height:auto;" /></center>
 More architecture details are shown below:  
 
 |Model Arch Component | Value |  
@@ -49,11 +49,15 @@ We encode protein sequence with single amino acid resolution with 44 vocabularie
 We assess the advantages of pretraining AIDO.Protein 16B through experiments across more than 300 tasks from two important protein benchmarks, xTrimoPGLM benchmark and ProteinGym DMS benchmark, encompassing residue-level, sequence-level, and protein-protein interaction (PPI) level tasks. We further adapted our model for structure-conditioned protein sequence generation tasks
 
 ## Results
-- xTrimoPGLM Benchmark
+### xTrimoPGLM Benchmark
+<center><img src="xtrimo_results.png" alt="An Overview of AIDO.Protein" style="width:70%; height:auto;" /></center>
 
-- ProteinGym DMS Benchmark
+### ProteinGym DMS Benchmark
+<center><img src="dms_results.png" alt="An Overview of AIDO.Protein" style="width:70%; height:auto;" /></center>
+
+### Inverse Folding Generation
+<center><img src="inverse_folding.png" alt="An Overview of AIDO.Protein" style="width:70%; height:auto;" /></center>
 
-- Inverse Folding Generation
 
 ## How to Use
 ### Build any downstream models from this backbone