fix explanation for v2 treatment.

app.py

@@ -164,8 +164,8 @@ with gr.Blocks(title="OCR QA Demo") as demo:
         - Note: Non-alphabetic characters will be removed. For efficiency reasons, all digits are replace by the digit 0.
 
     #### ⚠️ Limitations:
-    - The wordlists are **not exhaustive**, particularly for **historical vocabulary**, **dialects**, or **named entities**.
-    - The method may fail to flag **short OCR artifacts** (e.g., 1–2 character noise) and **non-alphabetic symbols**.
+    - The lists of known words are **not exhaustive**, particularly for **historical vocabulary**, **Luxembourgish**, or **named entities**.
+    - The method may fail to flag **short OCR artifacts** (e.g., 1–3 character noise).
 
     As such, the score should be understood as a **heuristic indicator**, best used for:
     - Comparative assessments between OCR outputs