Group Messaging Prompts for Multi-User Generative Chat Context
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional generative models are designed for single-user interactions and cannot maintain conversation context in group conversations, preventing their incorporation into computer-implemented messaging applications that support group discussions.
Innovation Solution
Incorporating a generative model into a computer-implemented messaging application by providing a prompt that identifies the conversation as a group conversation and includes participant identities and previous messages, allowing the model to generate outputs based on multiple user inputs.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If a conventional generative model is used for single-user interactions, then the model can maintain conversation context effectively, but the model cannot be incorporated into group conversation messaging applications
Solution Approach 1:
The system segments the group conversation context by creating separate conversation histories for each user-bot interaction within the group chat. The bot receives and processes messages individually while maintaining separate context windows for each participant, allowing it to track who said what without confusion from other group members' contributions.
Solution Approach 2:
The system introduces an intermediary layer between the generative model and the group conversation data. This intermediary processes and structures the multi-user input before presenting it to the model, managing context disambiguation and ensuring the bot receives well-formed prompts that clearly identify the relevant conversation participants and their messages.
2Adaptability or versatility
If a generative model is incorporated into a group conversation messaging application, then various use cases such as summarization, answering questions, and translation become available, but the model must be modified to handle multiple user inputs
Solution Approach 1:
The system performs preliminary processing of group conversation data before presenting it to the generative model. It pre-structures the context by identifying relevant participants, organizing their messages chronologically, and preparing disambiguated prompts that reduce the computational burden on the model during actual conversation processing.
Solution Approach 2:
The system dynamically adjusts parameters such as context window size, temperature settings, and prompt formatting based on the specific interaction scenario. It modifies how conversation history is presented to the model depending on factors like the number of active participants, message length, and type of task being performed.
Data Source
AI summary
A computing system includes a processor; and memory storing instructions that, when executed by the processor, cause the processor to perform several acts. The acts include receiving multiple messages from multiple users in a messaging application that supports group conversations, where the multiple messages are included in a group conversation. The acts also include providing a prompt to a generative model, where the prompt includes the multiple messages. The acts additionally include receiving, from the generative model, an output generated by the generative model based upon the prompt and including the output as a turn in the group conversation.