prospective-memory

The paper proposes a zero-inference prospective term for personal memory retrieval that boosts memory items linked to future commitments without query-time computation. It shows improved recall on a synthetic task set and positions this as part of a layered architecture for proactive AI assistants.

Introduces PM-Bench, a text-based benchmark for evaluating prospective memory in LLM agents, inspired by cognitive science. Experiments show that even the best method (GPT-5.4 agent) achieves only 65.1% F1 score, indicating significant challenges in reliable intention execution.