Upload folder using huggingface_hub

README.md

@@ -32,37 +32,40 @@ dtype: float16
 
 ## Model Details
 
-The merged model combines the conversational question answering capabilities of Llama3-ChatQA-1.5-8B with the bilingual proficiency of Llama3-8B-Chinese-Chat. The former excels in retrieval-augmented generation (RAG) and conversational QA, while the latter is fine-tuned for Chinese and English interactions, enhancing its performance in multilingual contexts.
+The merged model combines the conversational question answering capabilities of Llama3-ChatQA-1.5 with the bilingual proficiency of Llama3-8B-Chinese-Chat. Llama3-ChatQA-1.5 is designed for conversational QA and retrieval-augmented generation, leveraging a rich dataset to enhance its performance in understanding and generating contextually relevant responses. On the other hand, Llama3-8B-Chinese-Chat is fine-tuned specifically for Chinese and English users, excelling in tasks such as roleplaying and tool usage.
 
 ## Description
 
-This model is designed to provide a seamless experience for users engaging in conversations that require both English and Chinese language understanding. By merging these two models, we aim to leverage the strengths of each, resulting in improved contextual understanding and response generation across diverse topics.
+This model aims to provide a seamless experience for users who require both English and Chinese language support in conversational contexts. By merging these two models, we enhance the ability to handle diverse queries, allowing for more nuanced and context-aware interactions. The model is particularly effective in scenarios where users switch between languages or require responses that incorporate cultural nuances from both English and Chinese contexts.
 
 ## Merge Hypothesis
 
-The hypothesis behind this merge is that combining the strengths of a model optimized for conversational QA with one that is finely tuned for Chinese language interactions will yield a model capable of handling a wider range of queries and providing more nuanced responses in both languages.
+The hypothesis behind this merge is that combining the strengths of both models will yield a more capable and flexible conversational agent. The Llama3-ChatQA-1.5 model's proficiency in QA tasks complements the Llama3-8B-Chinese-Chat's bilingual capabilities, resulting in a model that can effectively engage users in both languages.
 
 ## Use Cases
 
-- **Conversational Agents**: Ideal for chatbots that need to respond to user queries in both English and Chinese.
-- **Multilingual Support**: Useful in applications requiring bilingual capabilities, such as customer support or educational tools.
-- **Content Generation**: Can be employed to generate text in both languages, catering to diverse audiences.
+- **Bilingual Customer Support**: Providing assistance in both English and Chinese for customer inquiries.
+- **Language Learning**: Assisting learners in practicing conversational skills in both languages.
+- **Cultural Exchange**: Facilitating discussions that require understanding of cultural references in both English and Chinese.
 
 ## Model Features
 
-The merged model benefits from:
-- Enhanced conversational capabilities, particularly in question answering.
-- Improved bilingual performance, allowing for fluid transitions between English and Chinese.
-- A comprehensive understanding of context, enabling more relevant and accurate responses.
+- **Bilingual Proficiency**: Capable of understanding and generating text in both English and Chinese.
+- **Conversational QA**: Enhanced ability to answer questions based on context, leveraging retrieval-augmented generation techniques.
+- **Roleplaying and Tool Usage**: Supports interactive scenarios where users can engage in roleplay or request tool-based assistance.
 
 ## Evaluation Results
 
-The evaluation results of the parent models indicate strong performance in their respective domains. For instance, Llama3-ChatQA-1.5-8B has shown significant improvements in conversational QA tasks, while Llama3-8B-Chinese-Chat has excelled in generating coherent and contextually appropriate responses in Chinese.
+The performance of the parent models in various benchmarks indicates strong capabilities in their respective domains. For instance, Llama3-ChatQA-1.5 has shown impressive results in conversational QA tasks, while Llama3-8B-Chinese-Chat has excelled in bilingual interactions, surpassing previous models in specific benchmarks.
+
+| Benchmark | Llama3-ChatQA-1.5-8B | Llama3-8B-Chinese-Chat |
+|-----------|------------------------|-------------------------|
+| Doc2Dial  | 41.26                  | N/A                     |
+| QuAC      | 38.82                  | N/A                     |
+| Average    | 58.25                  | N/A                     |
 
 ## Limitations of Merged Model
 
-While the merged model offers enhanced capabilities, it may still inherit some limitations from its parent models. Potential issues include:
-- Biases present in the training data of both models, which could affect the fairness and neutrality of responses.
-- Challenges in maintaining context when switching between languages, which may lead to occasional misunderstandings or inaccuracies.
+While the merged model offers enhanced capabilities, it may still inherit some limitations from the parent models. Potential biases in language understanding, particularly in cultural contexts, may affect the quality of responses. Additionally, the model's performance may vary based on the complexity of the queries and the context provided.
 
-In summary, Llama3-ChatQA-1.5-8B-Llama3-8B-Chinese-Chat-linear-merge represents a significant step forward in creating a bilingual conversational AI, combining the strengths of its predecessors while also acknowledging the challenges that come with such a complex integration.
+In summary, Llama3-ChatQA-1.5-Llama3-8B-Chinese-Chat-linear-merge represents a significant step towards creating a more inclusive and capable conversational AI, bridging the gap between English and Chinese interactions.