Chronos: Temporal-Aware Conversational Agents with Structured Event Retrieval for Long-Term Memory

2026-03-17

AI Summary

Existing dialogue systems struggle to effectively model time-evolving facts and user preferences over extended multi-turn interactions and lack efficient retrieval mechanisms for time-sensitive, multi-hop queries. This work proposes a novel approach that parses dialogue utterances into subject-predicate-object event triples annotated with temporal scopes and entity aliases, constructing both a structured event calendar and a contextual turn calendar. The method employs dynamic prompting to guide the model in temporal filtering and multi-hop reasoning. It introduces, for the first time, a structured event calendar coupled with a dynamic retrieval-guided mechanism, achieving 92.60% accuracy on the Chronos Low setting and 95.60% on Chronos High of the LongMemEvalS benchmark—surpassing the previous state-of-the-art system by 7.67%. Ablation studies indicate that the event calendar alone accounts for 58.9% of the overall performance gain.

Technology Category

• Data Mining & Knowledge Management: Conversational Systems for Recommendation & Retrieval
• Planning, Routing, and Scheduling: Planning with Language Models
• Natural Language Processing: Language Grounding & Multi-modal NLP

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

Recent advances in Large Language Models (LLMs) have enabled conversational AI agents to engage in extended multi-turn interactions spanning weeks or months. However, existing memory systems struggle to reason over temporally grounded facts and preferences that evolve across months of interaction and lack effective retrieval strategies for multi-hop, time-sensitive queries over long dialogue histories. We introduce Chronos, a novel temporal-aware memory framework that decomposes raw dialogue into subject-verb-object event tuples with resolved datetime ranges and entity aliases, indexing them in a structured event calendar alongside a turn calendar that preserves full conversational context. At query time, Chronos applies dynamic prompting to generate tailored retrieval guidance for each question, directing the agent on what to retrieve, how to filter across time ranges, and how to approach multi-hop reasoning through an iterative tool-calling loop over both calendars. We evaluate Chronos with 8 LLMs, both open-source and closed-source, on the LongMemEvalS benchmark comprising 500 questions spanning six categories of dialogue history tasks. Chronos Low achieves 92.60% and Chronos High scores 95.60% accuracy, setting a new state of the art with an improvement of 7.67% over the best prior system. Ablation results reveal the events calendar accounts for a 58.9% gain on the baseline while all other components yield improvements between 15.5% and 22.3%. Notably, Chronos Low alone surpasses prior approaches evaluated under their strongest model configurations.

Problem

• long-term memory
• temporal reasoning
• conversational agents
• time-sensitive retrieval
• multi-hop queries

Innovation

• temporal-aware memory
• structured event retrieval
• multi-hop reasoning
• dynamic prompting
• long-term dialogue history