Update README.md

README.md

@@ -63,8 +63,4 @@ Outputs:
 
 ## Training Data
 
-OSOR is trained on CORNE, a SAVP-verified effect-aware object-removal dataset. Evaluation uses CORNE-Val, AnimeEraseBench, TextEraseBench, and additional paired-background object-removal benchmarks.
-
-## Limitations
-
-OSOR may fail when the target object is extremely large, the mask is severely incorrect, or the background requires ambiguous semantic hallucination. As with other generative editing models, outputs should be reviewed before use in high-stakes contexts.
+OSOR is trained on CORNE, a SAVP-verified effect-aware object-removal dataset. Evaluation uses CORNE-Val, AnimeEraseBench, TextEraseBench, and additional paired-background object-removal benchmarks.