diff --git a/lambench/metrics/results/README.md b/lambench/metrics/results/README.md
index 19b3b3f..72c791f 100644
--- a/lambench/metrics/results/README.md
+++ b/lambench/metrics/results/README.md
@@ -19,7 +19,7 @@ Large atomistic models (LAM), also known as machine learning interatomic potenti
 The following changes have been made compared to the previouly release version v0.3.1:
 - Added new models: MACE-MH-1, DPA-3.2-5M
 - Updated `Force Field Prediction` tasks, and for the domain of `Molecules`, two sets of labels were provided to support OMol25-trained models.
-- Added new `Property Calculation` tasks: oxygen vacancy formation energy prediction, protein-ligand binding energy prediction, reaction energy barrier prediction, and volume prediction from materials under pressure.
+- Added new `Property Calculation` tasks: oxygen vacancy formation energy prediction, protein-ligand binding energy prediction, reaction energy barrier prediction, stacking fault energy prediction, and volume prediction from materials under pressure.
 
 <span style="color:red">⚠️ Note: To assess full LAM capacity, we use OMat24-trained task heads for *Force Field Prediction* in Inorganic Materials and Catalysis, and OMol25-trained task heads for Molecules, when available. As for *Property Calculation*, we follow a similar approach, but use OC20-trained task heads for Catalysis when available, as this tends to yield better performance.</span>
 
diff --git a/lambench/tasks/calculator/stacking_fault/stacking_fault.py b/lambench/tasks/calculator/stacking_fault/stacking_fault.py
index 322181e..9ae1fff 100644
--- a/lambench/tasks/calculator/stacking_fault/stacking_fault.py
+++ b/lambench/tasks/calculator/stacking_fault/stacking_fault.py
@@ -52,8 +52,8 @@ def calc_one_traj(traj, label, calc):
 def run_inference(model: ASEModel, test_data: Path) -> dict:
     calc = model.calc
 
-    traj_files = sorted(list(test_data.glob("*.traj")))
-    label_files = sorted(list(test_data.glob("*.csv")))
+    traj_files = sorted(list(test_data.rglob("*.traj")))
+    label_files = sorted(list(test_data.rglob("*.csv")))
 
     energy_maes = []
     derivative_maes = []
@@ -68,5 +68,5 @@ def run_inference(model: ASEModel, test_data: Path) -> dict:
 
     return {
         "MAE_E": np.round(np.mean(energy_maes), 4),  # mJ/m²
-        "MAE_dE": np.round(np.mean(derivative_maes), 4),  # mJ/m²/unit displacement
+        "MAE_dE": np.round(np.mean(derivative_maes), 4),  # 1/unit displacement
     }