Update README.md

README.md

@@ -1,32 +1,32 @@
 ---
 license: mit
 language:
-  - en
-  - ar
-  - bn
-  - de
-  - es
-  - hi
-  - ro
-  - ru
-  - zh
+- en
+- ar
+- bn
+- de
+- es
+- hi
+- ro
+- ru
+- zh
 base_model: Qwen/Qwen3-VL-8B-Instruct
 library_name: transformers
 pipeline_tag: image-text-to-text
 tags:
-  - memes
-  - multimodal
-  - multilingual
-  - hate-speech
-  - vision-language-model
-  - qwen3-vl
+- memes
+- multimodal
+- multilingual
+- hate-speech
+- vision-language-model
+- qwen3-vl
 datasets:
-  - QCRI/MemeLens-VLM
+- QCRI/MemeLens
 ---
 
 # MemeLens-VLM
 
-**MemeLens** is a unified multilingual, multitask Vision-Language Model (VLM) for meme understanding. It is fine-tuned from [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) using a **classify-then-explain** training strategy on the [MemeLens dataset](https://huggingface.co/datasets/QCRI/MemeLens-VLM), which consolidates 38 public meme datasets across 20 tasks and 9 languages.
+**MemeLens** is a unified multilingual, multitask Vision-Language Model (VLM) for meme understanding. It is fine-tuned from [Qwen3-VL-8B-Instruct](https://huggingface.co/Qwen/Qwen3-VL-8B-Instruct) using a **classify-then-explain** training strategy on the [MemeLens dataset](https://huggingface.co/datasets/QCRI/MemeLens), which consolidates 38 public meme datasets across 20 tasks and 9 languages.
 
 **Paper:** [MemeLens: Multilingual Multitask VLMs for Memes](https://arxiv.org/abs/2601.12539)
 
@@ -469,4 +469,4 @@ Explanation: <explanation>
       primaryClass={cs.AI},
       url={https://arxiv.org/abs/2601.12539},
 }
-```
+```