A cross-platform AI conversation context automatic injection system and method

Abstract

The application discloses a cross-platform AI conversation context automatic injection system and method, belongs to the field of artificial intelligence, and retrieves relevant historical memories from a separable memory storage carrier through a semantic matching module, automatically splices the relevant historical memories to the front end of current conversation input, and injects the relevant historical memories into different AI platforms through a cross-platform adaptation module, so that the cross-platform conversation context is coherent, the memory island problem of different AI platforms is solved, and users do not need to repeatedly explain the context, and the AI use experience is improved. The application is based on a separable memory storage architecture, and has the advantages of controllable privacy, non-sensing experience, strong compatibility and the like.

Description

Technical Field

[0001] This invention belongs to the fields of artificial intelligence, data storage, and natural language processing, and specifically relates to a cross-platform AI dialogue context automatic injection system and method. Background Technology

[0002] Current AI interaction systems have the following core flaws: 1. Cross-platform memory silos: Users' conversation memories on different AI platforms (OpenClaw / Doubao / ChatGPT / Claude, etc.) are independent and cannot be shared. 2. Inconsistent Context: When the same user uses AI on different platforms and devices, they need to repeatedly explain the historical context to the AI, resulting in a very poor user experience. 3. Low memory reuse rate: AI on different platforms cannot reuse users' historical memories, and the AI ​​needs to be retrained for each conversation. 4. High user barrier: When users use a new AI platform, they cannot quickly transfer their historical memories, and the learning cost is extremely high.

[0003] There is no effective solution to this problem in the existing technology, and users can only manually copy and paste historical conversations, which is extremely inefficient. Summary of the Invention

[0004] Technical Problem: The technical problem to be solved by this invention is to overcome the shortcomings of the prior art and provide a cross-platform AI dialogue context automatic injection system and method, so as to realize memory interoperability between different AI platforms, so that users do not need to repeatedly explain the context and improve the AI ​​user experience.

[0005] Technical solution: This invention discloses a cross-platform AI dialogue context automatic injection system, characterized in that it includes: (1) Semantic matching module: Real-time monitoring of the user's current dialogue semantics, and retrieval of relevant historical memories from the separable memory storage carrier; (2) Context splicing module: Based on the semantic matching results, it automatically splices relevant historical memories to the beginning of the user's current input to form a complete dialogue context; (3) Cross-platform adaptation module: compatible with the dialogue format of all mainstream AI platforms, automatically adapts to the input rules of different AI platforms, and automatically injects them into AI dialogue; (4) Access control module: supports users to customize which memories can be injected into AI and which private memories should not be injected; (5) Real-time synchronization module: Real-time synchronization of dialogue between different platforms to a separable memory storage medium.

[0006] This invention also discloses a cross-platform AI dialogue context automatic injection method, comprising the following steps: S1: Listen to the user's current dialogue input on the AI ​​platform; S2: Perform semantic analysis on the current dialogue input to extract core keywords and semantic features; S3: Retrieve the N most similar historical memories from separable memory storage carriers based on semantic features; S4: Sort the retrieved historical memories according to time order and semantic relevance, and automatically append them to the beginning of the current dialogue input to form a complete context-enhanced dialogue input; S5: The concatenated dialogue input is automatically submitted to the AI ​​platform, which then generates a response based on the complete context. S6: Automatically save the content of this conversation to a detachable memory storage medium and update the memory bank.

[0007] Beneficial effects: 1. Cross-platform coherence: Users can automatically inherit all historical memories in conversations on any AI platform, eliminating the need for repeated explanations of context. 2. Seamless experience: The user is completely unaware of the entire process, just like a normal conversation. 3. Privacy Control: Users have complete control over which memories can be accessed by AI, and privacy data will not be leaked. 4. Strong compatibility: It is based entirely on existing separable memory storage systems, requiring no additional hardware and only software upgrades. Attached Figure Description Figure 1 This is a schematic diagram of the system architecture of an embodiment of the present invention. Figure 2 This is a schematic diagram of the method flow according to an embodiment of the present invention. Detailed Implementation

[0008] The cross-platform AI dialogue context automatic injection system of the present invention is based on an existing separable encrypted memory storage system. Users store memory data through a pluggable hardware storage medium and automatically synchronize dialogue records of different AI platforms through browser plugins or desktop clients. The semantic matching module uses the BERT vector similarity algorithm to retrieve relevant historical memories and automatically splices them into the user's current dialogue input front end to achieve cross-platform context coherence.

Claims

1. A cross-platform AI dialogue context automatic injection system, characterized in that, include: The semantic matching module is used to perform semantic analysis on the user's current dialogue input and retrieve relevant historical memories from a separable memory storage carrier independent of the smart terminal itself. The context splicing module is used to automatically splice the retrieved relevant historical memories to the front end of the current dialogue input, forming a complete dialogue context that includes the combination of historical memories and the current input; The cross-platform adaptation module is used to automatically inject the spliced ​​complete dialogue context into the dialogue interface of different AI platforms to achieve cross-platform dialogue context coherence. The separable memory storage carrier is a hardware carrier module in the separable encrypted memory storage and retrieval system.

2. The system according to claim 1, characterized in that, The semantic matching module uses a vector similarity matching algorithm to calculate the semantic similarity between the current dialogue input and historical memories, and returns the Top-K most relevant historical memories.

3. The system according to claim 1, characterized in that, The context splicing module supports custom splicing rules, allowing users to customize historical memory length, relevance threshold, and privacy filtering rules.

4. The system according to claim 1, characterized in that, The cross-platform adaptation module supports three adaptation methods—browser plugins, desktop clients, and mobile SDKs—to achieve context injection, ensuring compatibility with all mainstream AI platforms.

5. The system according to claim 1, characterized in that, It also includes an access control module, allowing users to set which types of memories can be injected into AI and which private memories are prohibited from being injected.

6. The system according to claim 1, characterized in that, It also includes a real-time synchronization module, which automatically synchronizes and saves the current conversation content to a detachable memory storage medium.

7. A cross-platform AI dialogue context automatic injection method, characterized in that, The system described in any one of claims 1-6 is used to realize the function of automatic injection of cross-platform AI dialogue context.

Images