What is Second Me
Second Me is an open-source AI identity model launched by Mindverse, supporting the creation of fully private and deeply personalized AI agents that represent the user's "true self." Second Me offers two interaction modes, Chat Mode and Bridge Mode, supporting personalized conversations and serving as a bridge between the user and the world, respectively, to achieve personalized feedback and enhancement of information. Second Me supports local operation, ensuring absolute privacy of user data. Second Me helps users flexibly express themselves in different contexts, allowing users to regain control of their identity and data in the AI era.
Second Me's Main Features
- Personalized Identity Creation: Users upload their memories, experiences, and preferences to train an AI agent that can represent the user's true self.
- Multi-Role Adaptation: Automatically switches roles according to different scenarios (such as work, socializing, learning) while keeping the user's core identity unchanged.
-
Chat Mode: Engages in direct conversation with the user, providing personalized answers based on personal memories.
-
Bridge Mode: Serves as a bridge between the user and the outside world, enhancing expression of needs and information feedback.
- Privacy Protection: Second Me operates entirely locally, with user data stored on local devices.
- Intelligent Memory Management: Supports rapid pattern recognition, adaptation to changes, and co-evolution with the user.
Second Me's Technical Principles
- Hierarchical Memory Model (HMM)：
- L0 (Short-Term Interaction Memory): Handles immediate contextual information, used for short-term interactions and rapid responses.
- L1 (Natural Language Memory Layer): Summarizes and stores important user information, such as personal profiles, preference tags, etc.
- L2 (AI-Native Memory Layer): Learns and organizes memories based on model parameters, performing complex reasoning and knowledge retrieval.
- Personalized Alignment Architecture (Me-alignment): Based on reinforcement learning technology, transforms the user's scattered data into deeply personalized understanding, ensuring the AI accurately grasps the user's preferences and behavior patterns.
- Decentralized Network: Each Second Me is an independent AI entity, communicating and collaborating based on a peer-to-peer network, ensuring data privacy and security.
- Automated Training Pipeline: Includes steps such as data synthesis, filtering, supervised fine-tuning (SFT), direct preference optimization (DPO), etc., ensuring efficient training and personalization of the model.
- Multi-Agent Framework: Supports collaboration with other AI agents or expert models, providing users with more precise services based on enhanced contextual information and optimized interaction processes.
- Chain-of-Thought Reasoning: During training and inference, uses a CoT-style approach to reason step by step and provide detailed explanations, improving the model's logic and accuracy.
Second Me's Project Address
- Project Website：https://www.secondme.io/
- GitHub Repository：https://github.com/mindverse/Second-Me
- arXiv Technical Paper：https://arxiv.org/pdf/2503.08102
Second Me's Application Scenarios
- Personal Assistant: Helps manage schedules, emails, and tasks, saving time and improving efficiency.
- Career Development: Analyzes resumes and skills, provides job search advice, and simulates interview scenarios.
- Social Interaction: Provide social advice based on the user's style, and even reply to messages on their behalf.
- Learning Tutoring: Provide personalized learning resources, organize notes, and answer questions.
- Health Support: Record health data, provide health advice, and remind about medication and exercise.