diff --git a/openseek/competition/LongContext-ICL-Annotation/data/README.md b/openseek/competition/LongContext-ICL-Annotation/data/README.md
index f30f56b..5535c4b 100644
--- a/openseek/competition/LongContext-ICL-Annotation/data/README.md
+++ b/openseek/competition/LongContext-ICL-Annotation/data/README.md
@@ -21,7 +21,7 @@ The datasets are specifically designed to evaluate the capability of Large Langu
 | openseek-5 | semeval_2018_task1_tweet_sadness_detection |  30K  | 500 | 
 | openseek-6 | mnli_same_genre_classification |  30K  | 500 | 
 | openseek-7 | jeopardy_answer_generation_all |  30K  | 500 | 
-| openseek-8 | kernel_genernation |  15K  | 166 | 
+| openseek-8 | kernel_genernation |  16K  | 166 | 
 
 
 ## Data Structure
diff --git a/openseek/competition/LongContext-ICL-Annotation/data/README_zh.md b/openseek/competition/LongContext-ICL-Annotation/data/README_zh.md
index cc565fe..d83213a 100644
--- a/openseek/competition/LongContext-ICL-Annotation/data/README_zh.md
+++ b/openseek/competition/LongContext-ICL-Annotation/data/README_zh.md
@@ -22,7 +22,7 @@
 | openseek-5 | semeval_2018_task1_tweet_sadness_detection |  30K  | 500 | 
 | openseek-6 | mnli_same_genre_classification |  30K  | 500 | 
 | openseek-7 | jeopardy_answer_generation_all |  30K  | 500 | 
-| openseek-8 | kernel_genernation |  15K  | 166 | 
+| openseek-8 | kernel_genernation |  16K  | 166 | 
 
 ---