Dhruv Dubey

Token-Efficient Conversational History Management using Sparse Semantic Patch Memory

2026 · Zenodo (CERN European Organization for Nuclear Research)

Updated version of the Sparse Semantic Patch Memory (SSPM) framework including full architecture, Gemini 2.5 Flash based experimental validation, and extended empirical evaluation demonstrating significant token reduction while preserving reasoning constraints.Large Language Model (LLM) deployments …