Update README.md

README.md

@@ -84,7 +84,7 @@ If you use this dataset in your work, please cite both the VocSim benchmark pape
 
 ```bib
 @unpublished{vocsim2025,
-  title={VocSim: Zero-Shot Audio Similarity Benchmark for Neural Embeddings},
+  title={VocSim: A Training-Free Benchmark for Content Identity in Single-Source Audio Embeddings},
   author={Anonymous},
   year={2025},
   note={Submitted manuscript}