Abstract
Current Large Language Model (LLM) interfaces often retain chat threads indefinitely or for extended fixed periods, leading to redundancy and management difficulties for users. This issue is particularly acute for time-bound or topic-specific queries where the relevance of the conversation diminishes shortly after the initial interaction.
A method is disclosed for the automated management of chat history through contextual analysis. Intra-thread and shared contexts are evaluated to determine the ongoing utility of specific conversations. Based on this analysis, threads are automatically deleted once time-sensitive information expires or merged when multiple threads address the same underlying topic. Users may also provide specific instructions within a prompt to trigger these organizational actions across their history. This technology improves workspace hygiene and reduces the manual effort required to organize large volumes of AI-generated interactions, ensuring the chat interface remains relevant to the user's current needs.
Creative Commons License
This work is licensed under a Creative Commons Attribution 4.0 License.
Recommended Citation
Lanning, Gabi; Boiarshinov, Dmitrii; and Guajardo, Jaime, "Automated Management of Large Language Model Chat Threads via Contextual Analysis", Technical Disclosure Commons, ()
https://www.tdcommons.org/dpubs_series/10107