Abstract
On-demand content summarization requests, e.g., for various types of content stored and handled by cloud productivity suites, can result in large wait times and strained resources due to the computational load imposed by content summarization models. Pre-generating content summaries can inefficiently duplicate efforts and generate content with low usage probability. This disclosure describes machine-learning techniques, implemented with user permissions for content access, for scalable and intelligent content pregeneration, e.g., in cloud productivity suites. Computational resources are allocated for content summarization and summarization models are selected based on content value, avoiding redundant pregeneration. The techniques advantageously reduce latency, optimize resource utilization, and improve content accessibility and comprehensibility, especially for complex or fanned-out content.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Start, Johannes Elias and Lunney, John James, "Scalable Pregeneration of Content Summaries", Technical Disclosure Commons, (June 10, 2025)
https://www.tdcommons.org/dpubs_series/8220