Qwen3-4B-customerservice-context-summarization-llm-judge-data

该数据集名为'customer-service-context-summarization-evaluation-data'，主要用于客户服务场景下的上下文摘要生成与评估。数据集包含1000个训练样本，总大小为4.5MB。每个样本包含多个字段：对话ID(conversation_id)、指令(instruction)、对话历史(conversation_history)、摘要提示(summarization_prompt)、历史摘要(history_summary)、客户问题(client_question)、客服回答(agent_answer)、优化后的客服回答(refined_agent_answer)、生成的摘要(generated_summary)以及三个评估指标：信息准确性(Information-Accuracy)、结构清晰度(Structural-Clarity)和忠实度(Faithfulness)。该数据集适用于自然语言处理任务，特别是对话摘要生成和质量评估领域。This dataset, named customer-service-context-summarization-evaluation-data, is primarily used for context summarization generation and evaluation in customer service scenarios. It contains 1000 training samples with a total size of 4.5 MB. Each sample includes multiple fields: conversation_id, instruction, conversation_history, summarization_prompt, history_summary, client_question, agent_answer, refined_agent_answer, generated_summary, and three evaluation metrics: Information-Accuracy, Structural-Clarity, and Faithfulness. This dataset is applicable to natural language processing (NLP) tasks, particularly in the domains of dialogue summarization generation and quality evaluation.