Update README.md

README.md

@@ -398,14 +398,9 @@ messages = [
 
 ## Limitations
 
-While Qwen2-VL are applicable to a wide range of visual tasks, it is equally important to understand its limitations. Here are some known restrictions:
-
-1. Lack of Audio Support: The current model does **not comprehend audio information** within videos.
-2. Data timeliness: Our image dataset is **updated until June 2023**, and information subsequent to this date may not be covered.
-3. Constraints in Individuals and Intellectual Property (IP): The model's capacity to recognize specific individuals or IPs is limited, potentially failing to comprehensively cover all well-known personalities or brands.
-4. Limited Capacity for Complex Instruction: When faced with intricate multi-step instructions, the model's understanding and execution capabilities require enhancement.
-5. Insufficient Counting Accuracy: Particularly in complex scenes, the accuracy of object counting is not high, necessitating further improvements.
-6. Weak Spatial Reasoning Skills: Especially in 3D spaces, the model's inference of object positional relationships is inadequate, making it difficult to precisely judge the relative positions of objects.
+- This model is starting from Qwen2-VL-7B-Base, so it may have limitations mentioned in https://huggingface.co/Qwen/Qwen2-VL-7B.
+- This model is trained only with streaming frame-words paradigm, thus it may be only capable for real-time video commentary.
+- The training ASR data is from YouTube CC, which has well-known low quality, so its formatting is not good (e.g. cannot output punctuation).
 
 These limitations serve as ongoing directions for model optimization and improvement, and we are committed to continually enhancing the model's performance and scope of application.