Year: 2026
Type: conference-paper
Abstract: Large language model agents increasingly rely on memory to support long-horizon interaction, yet existing frameworks expose only a small set of low-level primitives and lack a formal, executable specification for memory control.As a result, higher-order operations such as promotion, consolidation, and lifecycle governance are often missing or inconsistently implemented, leading to fragmented memory management across systems.We introduce Text2Mem 1 , a unified memory operation language for translating natural-language instructions into executable memory operations.Text2Mem defines a compact operation set spanning encoding, storage, and retrieval, and represents each instruction as a schema-based contract with explicit fields and semantic invariants.Validated schemas are parsed into typed operation objects and executed through a unified pipeline with a SQL-based reference backend, providing a controlled and auditable execution environment for memory operations.We further present the Text2Mem Benchmark, which decouples schema generation from execution to systematically evaluate planning accuracy and execution fidelity.Together, Text2Mem and its benchmark provide a standardized intermediate specification and evaluation foundation for studying controllable and reproducible memory management in LLM-based agents.
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
OpenAlex ID: W7166866487