iAgents - A multi-AI agent collaborative framework launched by Tsinghua University
iAgents is a multi-AI agent collaboration framework developed by Tsinghua University. It's based on equipping each user with a personal AI agent to facilitate collaboration and complete daily tasks. The agents can understand user files and commands, learn from collaboration, and achieve self-management...
What is iAgents?
iAgents is a multi-AI agent collaboration framework developed by Tsinghua University. It's based on equipping each user with a personal AI agent to facilitate collaboration and complete daily tasks. The agents can understand user files and commands, learn from collaboration, and achieve automated information exchange and task resolution. iAgents uses the infoNav inference mechanism to guide agents in effective communication even under information asymmetry, improving team collaboration efficiency and driving the cutting-edge development of AI technology. The system is completely open source and supports seamless integration with existing tools and platforms.
iAgents' main functions
- Information exchange and collaboration:The iAgents system enables intelligent agents to simulate information exchange in human social networks, solving the problem of information asymmetry among multiple agents.
- Task solving ability:The system can handle complex collaborative tasks, such as finding specific information in large social networks (Needle-Oriented tasks) and performing algorithmic reasoning based on distributed information (Reasoning-Oriented tasks).
- Autonomous communication:Intelligent agents can represent multiple human users in autonomous communication, completing multiple rounds of dialogue and task resolution.
- Information retrieval and memory:The system is designed with a hybrid memory mechanism, including clear memory and fuzzy memory, to store and retrieve human information.
iAgents' technical principles
- Information Navigation Mechanism (InfoNav)It helps intelligent agents plan and track the information they need in situations of information asymmetry, and guides them to know what to ask and what to answer.
- Hybrid memory mechanismiAgents designs a hybrid memory mechanism, including clear memory and fuzzy memory, to organize and retrieve human information.
- Clear Memory: Storing information in a structured format facilitates accurate retrieval.
- Fuzzy memoryStores a summary of the conversation text, retrieved using an embedding-based approximate nearest neighbor (ANN).
- Multi-agent collaborationiAgents simulates human social networks, enabling intelligent agents to communicate and exchange information autonomously to solve tasks.
- Natural Language ProcessingIntelligent agents can understand and process natural language, enabling them to communicate naturally with users.
iAgents project address
- Project official website:thinkwee.top/iagents
- GitHub repository:https://github.com/thinkwee/iagents
- arXiv technical paper:https://arxiv.org/pdf/2406.14928
Application scenarios of iAgents
- Meeting CoordinationThe intelligent fitness system automatically schedules and coordinates meetings on behalf of users, determines the availability of all participants, and automatically books meeting rooms.
- project managementIn project management, intelligent agents help assign tasks, track progress, and synchronize information among project members.
- Customer ServiceThe intelligent system can automatically process customer inquiries, provide instant feedback, and escalate issues to human intervention when necessary.
- Procurement and Supply Chain ManagementThe intelligent system can automatically contact suppliers to compare prices, coordinate purchase orders, and manage communication within the supply chain.
- R&D collaborationIn the field of research and development, intelligent agents help coordinate the work of different teams, such as the communication and iterative development between product managers and engineers.