TR-EduVSum: A Turkish-Focused Dataset and Consensus Framework for Educational Video Summarization
This work addresses the need for automated, reproducible summarization in Turkish educational content, with potential low-cost generalization to other Turkic languages, though it is incremental as it builds on existing pyramid-based approaches.
The study tackled the problem of generating gold-standard summaries for Turkish educational videos by creating the TR-EduVSum dataset with 82 videos and 3281 human summaries, and proposed the AutoMUP method that achieved high semantic overlap with robust LLM summaries like Flash 2.5 and GPT-5.1.
This study presents a framework for generating the gold-standard summary fully automatically and reproducibly based on multiple human summaries of Turkish educational videos. Within the scope of the study, a new dataset called TR-EduVSum was created, encompassing 82 Turkish course videos in the field of "Data Structures and Algorithms" and containing a total of 3281 independent human summaries. Inspired by existing pyramid-based evaluation approaches, the AutoMUP (Automatic Meaning Unit Pyramid) method is proposed, which extracts consensus-based content from multiple human summaries. AutoMUP clusters the meaning units extracted from human summaries using embedding, statistically models inter-participant agreement, and generates graded summaries based on consensus weight. In this framework, the gold summary corresponds to the highest-consensus AutoMUP configuration, constructed from the most frequently supported meaning units across human summaries. Experimental results show that AutoMUP summaries exhibit high semantic overlap with robust LLM (Large Language Model) summaries such as Flash 2.5 and GPT-5.1. Furthermore, ablation studies clearly demonstrate the decisive role of consensus weight and clustering in determining summary quality. The proposed approach can be generalized to other Turkic languages at low cost.