GraphRAG is a method that combines knowledge graphs with retrieval-augmented generation. It first extracts entities, relationships, and events from a ...
AI Guardrails, commonly translated as AI Guardrails, refer to constraints and detection mechanisms established around model input, context, tool calls...
DPO (Direct Preference Optimization) is a model alignment method. Training data is not a single standard answer but a pair of responses under the same...
Continuous Batch Processing is a dynamic scheduling method on the server side of large models: the system does not have to wait for all requests in th...
Speculative Decoding is a large model inference acceleration method: first, a lighter, faster draft model proposes several candidate tokens in success...
AI Evals refers to systematic evaluation of large models or AI applications. It's not just about asking a few random questions to get a feel, but turn...
LoRA stands for Low-Rank Adaptation, commonly referred to in Chinese as '低級適' (低級配). It is a highly efficient method for fine-tuning parameters: inste...
A vector database is a database specifically designed to store, index, and query vector data. Ordinary databases excel at precise field-based queries,...
