Update README.md

README.md

@@ -164,23 +164,23 @@ configs:
   - config_name: pbe
     data_files: 
     - split: train
-      path: MatPES-PBE-2025.2.json
+      path: MatPES-PBE-2025.2.jsonl
   - config_name: r2scan
     data_files: 
     - split: train
-      path: MatPES-R2SCAN-2025.2.json
+      path: MatPES-R2SCAN-2025.2.jsonl
   - config_name: pbe-2025.2
-    data_files: MatPES-PBE-2025.2.json
+    data_files: MatPES-PBE-2025.2.jsonl
   - config_name: r2scan-2025.2
-    data_files: MatPES-R2SCAN-2025.2.json
+    data_files: MatPES-R2SCAN-2025.2.jsonl
   - config_name: pbe-2025.1
-    data_files: MatPES-PBE-2025.1.json
+    data_files: MatPES-PBE-2025.1.jsonl
   - config_name: r2scan-2025.1
-    data_files: MatPES-R2SCAN-2025.1.json
+    data_files: MatPES-R2SCAN-2025.1.jsonl
   - config_name: pbe-atoms
-    data_files: MatPES-PBE-atoms.json
+    data_files: MatPES-PBE-atoms.jsonl
   - config_name: r2scan-atoms
-    data_files: MatPES-R2SCAN-atoms.json
+    data_files: MatPES-R2SCAN-atoms.jsonl
 papers:
   - 2503.04070
 ---
@@ -206,10 +206,12 @@ table.  MatPES is an initiative by the [Materialyze] Lab and the [Materials Proj
 3. **Quality.** MatPES includes computed data from the PBE functional, as well as the high fidelity r2SCAN meta-GGA
    functional with improved description across diverse bonding and chemistries.
 
-The initial v2025.1 release comprises ~400,000 structures from 300K MD simulations. This dataset is much smaller
-than other PES datasets in the literature and yet achieves comparable or, in some cases,
+The initial v2025.1 release comprises ~400,000 structures from 300K MD simulations. The v2025.2 removes some duplicate structures and adds charge information.
+This dataset is much smaller than other PES datasets in the literature and yet achieves comparable or, in some cases,
 [improved performance and reliability](http://matpes.ai/benchmarks) on trained FPs.
 
+The dataset is provided as jsonl files to facilitate memory-efficient streaming. The original json files are retained for backwards compatibility.
+
 MatPES is part of the MatML ecosystem, which includes the [MatGL] (Materials Graph Library) and [maml] (MAterials
 Machine Learning) packages, the [MatPES] (Materials Potential Energy Surface) dataset, and the [MatCalc] (Materials
 Calculator).