[공지사항]을 빙자한 안부와 근황 
Show more

Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

LearnLens: LLM-Enabled Personalized, Curriculum-Grounded Feedback with Educators in the Loop

Created by
  • Haebom

Author

Runcong Zhao, Artem Bobrov, Jiazheng Li, Yulan He

Outline

LearnLens is a modular LLM-based system for generating personalized, curriculum-aligned feedback in science education. It consists of three components: an error-aware assessment module that catches subtle inference errors; a curriculum-based generation module that uses structured topic-linked memory chains instead of traditional similarity-based retrieval to increase relevance and reduce noise; and a teacher engagement interface for customization and supervision. LearnLens addresses key challenges in existing systems to provide scalable, high-quality feedback that is effective for both teachers and students.

Takeaways, Limitations

•
Takeaways:
◦
Presenting new possibilities for providing efficient and personalized feedback in science education
◦
Reduce teachers' workload by utilizing LLM-based system
◦
Increase learning effectiveness by creating structured feedback linked to the curriculum
◦
Continuous improvement and customization of the system through teacher participation
•
Limitations:
◦
The performance of the system may depend on the quality of the LLM and course data used.
◦
Possible limitations in accurately capturing and interpreting subtle inference errors
◦
The ethical and bias issues of LLM-based systems need to be considered
◦
Additional research and verification needed for application in actual educational settings
👍
Made with Slashpage