A post-retrieval temporal layer designed to improve RAG systems by addressing time-blindness in vector searches. This library implements validity filtering, document kind classification, and exponential decay scoring to ensure retrieved information is fresh and accurate. It functions downstream of existing vector search systems without requiring re-indexing or new infrastructure.
>"How I added temporal awareness and freshness tracking to a RAG system with no sense of time."
An exploration of how AI models can be trained to understand and incorporate the concept of time, its passage, and its significance in various contexts.
An MCP server that gives language models temporal awareness and time calculation abilities. Teaching AI the significance of the passage of time through collaborative tool development.
LLMs as cognitive archaeologists, excavating the lost civilizations of thought. The article discusses how large language models preserve fragments of human cognition, the collapse of time within them, the persistence of bias and outdated ideas, and the need for human curation of AI's memory.
