高中语文主观题

该数据集聚焦于高中区域联考、学校期中期末考试中的关键主观题领域，涵盖 “文学类阅读”“文言文阅读”“古代诗歌阅读”“语言文字应用 1”“语言文字应用 2” 等五类重要题型。每类题型均配备充足且丰富的训练数据，题目来源广泛且具有代表性。数据构成包含清晰的题目、学生回答以及精准的评分信息。在题目方面，每类题型涵盖一百个以上风格各异、难度有别的题目，全面考查学生对高中语文知识的掌握与运用能力。学生回答质量呈现多元化，各分数段均有覆盖，尤其注重高低分段数据的收集，占比均不少于 10%，确保模型能充分学习不同水平学生的作答特征。学校来源广泛，全面覆盖当地教学水平处于前、中、后的各类学校，使数据更具普遍性与参考性。此数据集可广泛应用于教育大模型训练，助力教育局精准决策、优化中小学作业与考试设计，还能辅助教师智能评分，为学生提供个性化写作辅导，推动教育向智能化、高效化迈进。① 主观题数据产品由“错题本”服务中，学生提供的答题卡进行匿名化加工后形成，不含个人信息。试题产品由“错题本”服务中老师上传的试题加工形成。②所附隐私政策在用户注册登录APP时由用户点击确认接受，已提示未成年人在成年人陪同下阅读。数据授权政策出现的场景是用户申请使用“错题本”功能时，试题和答题卡数据的加工使用已获完整授权。

This dataset focuses on key subjective questions from high school regional joint exams, mid-term and final-term school exams, covering five major question types: "Literature Reading", "Classical Chinese Reading", "Ancient Poetry Reading", "Language Application 1" and "Language Application 2". Each question type is equipped with sufficient and rich training data, with widely sourced and representative questions. The data structure includes clear questions, student responses and accurate scoring information. Each question type includes more than 100 questions with varying styles and difficulty levels, comprehensively assessing students' mastery and application of high school Chinese knowledge. The quality of student responses is diverse, covering all score ranges, with special emphasis on the collection of high and low score segments, each accounting for no less than 10%, ensuring that the model can fully learn the response characteristics of students at different proficiency levels. Schools are widely sourced, comprehensively covering all local schools with teaching levels at top, middle and bottom tiers, making the data more universal and referential. This dataset can be widely used for training educational large models, helping education bureaus make precise decisions, optimize the design of primary and secondary school homework and exams, assist teachers in intelligent scoring, provide personalized writing tutoring for students, and promote the intelligent and efficient development of education. ① The subjective question data products are formed after anonymization processing of the answer sheets provided by students in the "Error Correction Notebook" service, and do not contain any personal information. The test question products are formed by processing the test questions uploaded by teachers in the "Error Correction Notebook" service. ② The attached privacy policy is confirmed and accepted by users when they register and log in to the APP, and minors have been prompted to read it accompanied by adults. The data authorization policy is presented when users apply for the "Error Correction Notebook" function, and the processing and use of test questions and answer sheet data have obtained complete authorization.