PEARL: Self-Evolving Assistant for Time Management with Reinforcement Learning

2026-01-17

AI Summary

This work addresses preference-driven decision-making in calendar conflicts—such as choosing to attend, reschedule, or decline overlapping meetings—by proposing PEARL, a reinforcement learning–based self-evolving language agent. PEARL incorporates an external memory module and a round-wise dynamic reward mechanism to enable continuous online modeling and adaptation to user preferences. As the first reinforcement learning framework specifically designed for calendar conflict resolution, PEARL significantly outperforms existing baselines on the newly introduced CalConflictBench benchmark, reducing the average error rate by 55% and achieving an error reduction ratio of 0.76, thereby demonstrating its effectiveness and generalization capability in long-horizon scenarios.

Technology Category

• Planning, Routing, and Scheduling: Planning with Language Models
• Multiagent Systems: Adversarial Agents
• Humans and AI: Human-Aware Planning and Behavior Prediction

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation
• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions

Abstract

Overlapping calendar invitations force busy professionals to repeatedly decide which meetings to attend, reschedule, or decline. We refer to this preference-driven decision process as calendar conflict resolution. Automating this decision process is crucial yet challenging. Scheduling logistics can drain hours, and human delegation often fails at scale, which motivates us to ask: Can we trust large language models (LLMs) or language agents to manage time? To enable a systematic study of this question, we introduce CalConflictBench, a benchmark for long-horizon calendar conflict resolution. In CalConflictBench, conflicts are presented to agents round-by-round over a calendar year, requiring them to infer and adapt to user preferences progressively. Our experiments show that current LLM agents perform poorly with high error rates, e.g., Qwen-3-30B-Think has an average error rate of 35%. To address this gap, we propose PEARL, a reinforcement-learning framework that (i) augments the language agent with an external preference memory that stores and updates inferred strategies (e.g., attendee priorities, topic importance, time/location preferences), and (ii) optimizes the agent with round-wise rewards that directly supervise decision correctness, ranking quality, and memory usage across rounds. Experiments on CalConflictBench show that PEARL achieves an error reduction rate of 0.76 and a 55% improvement in average error rate compared to the strongest baseline.

Problem

• calendar conflict resolution
• time management
• preference-driven decision
• large language model
• scheduling automation

Innovation

• reinforcement learning
• language agent
• calendar conflict resolution
• preference adaptation
• external memory