mlx-community/tnc-archive

该数据集是从“新研究中心与实践”的“非会员可用”数据中抓取的标题和摘要描述，数据来源于其网站。数据集100%由人类制作，每个描述都是由负责并交付材料或为客座讲师策划课程的讲师制作的。计划中还包括扩展，如视频会话的转录。数据集适用于多种下游任务，如训练图书馆员、生成教学大纲、作为RAG的参考、少样本学习等。数据集结构包括巴洛克风格的标题和100%人工编写的描述，大小不一，大多数介绍主题、每个研讨会会议的摘要，有时包括阅读材料。数据集的局限性在于其小众主题和对研讨会实际内容的受限访问。

Scraped titles and summarized descriptions of the "non-members available" data of the Seminars of The New Centre for Research & Practice, took this from our website where i have a status of god of FireStoreStoNe (FSSN). 100% Human-made, and not "just" humans, but each of the descriptions is made by Instructor, a person that was responsible for and delivered the materials themselves or curated the sessions for the guest lecturers. In plans only expansion, I want to persevere with transcripts of video sessions, after all, TNC plans training a model of their own. Small? Yes, but academic language quality steps up where quantity fails...! Downstream tasks include and are not limited to: training librarian from archives department-assistent or agent; generating syllabuses and other educational metadata generations; references for RAG and/or some figures of the leading avantgarde in SoTA thought, including the experimental genres of thought such as theory-fiction; few-shot learning; usable for training for analogous cases when you need to create a database for niche or NietzSHe subject. Baroque Titles & 100% human-written Descriptions. Sizes vary, most of them introduce the topics, abstracts of each Seminars Session, sometimes include Readings. Main limitation is the niche topic and the gated access to the actual content of the seminar.