Traj-Evolve: A Self-Evolving Multi-Agent System for Patient Trajectory Modeling in Lung Cancer Early Detection
Quick Take
Traj-Evolve is a self-evolving multi-agent system that enhances lung cancer early detection by modeling patient trajectories using longitudinal EHRs. It outperforms nine baselines, particularly in never-smoker populations, by integrating an Experience Pool and multi-agent reinforcement learning for optimized patient context retrieval and collaboration.
Key Points
- Utilizes an Experience Pool for non-parametric memory and few-shot context retrieval.
- Employs multi-agent reinforcement learning to optimize collaboration between agents.
- Achieves superior performance on lung cancer prediction tasks over nine strong baselines.
- Demonstrates improved specificity and sensitivity in risk prediction through complementary mechanisms.
- Highlights the importance of verified patient data for enhancing temporal reasoning in agents.
Article Content
From source RSS / original summaryarXiv:2606. 02812v1 Announce Type: new Abstract: Modeling patient trajectories from longitudinal electronic health records (EHRs) requires reasoning over sparse, noisy, and long-context multimodal sequences. Existing LLM-based multi-agent systems address context length but process patients in isolation, failing to mirror how clinicians leverage accumulated experience from similar prior cases. We present Traj-Evolve, a self-evolving multi-agent system with two complementary evolving mechanisms.
First, an Experience Pool (ExPool) acts as a non-parametric memory, indexing rejection-sampled reasoning traces to retrieve similar patients as few-shot contexts. Second, multi-agent reinforcement learning (MARL) via reward-ranked fine-tuning parametrically optimizes inter-agent and agent-memory collaboration. A leave-one-out cross-retrieval strategy unifies the two, aligning training- and inference-time behavior under retrieval augmentation.
On a lung cancer prediction task utilizing up to five years of multimodal EHRs, Traj-Evolve outperforms 9 strong baselines on the overall population and a challenging never-smoker population.
Analysis of the evolving dynamics highlights three key findings: (1) expanding the ExPool shifts optimal retrieval from diverse to specific samples; (2) under MARL, the manager agent's prediction loss converges quickly while the worker agents' temporal reasoning continues to benefit from more verified patients; and (3) the two mechanisms are complementary on the predicted risk, where ExPool improves specificity while MARL improves sensitivity.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.AI
See more →AUDITFLOW: Executable Symbolic Environments for Structured Financial Reporting Verification
AuditFlow introduces a multi-agent framework for structured financial reporting verification, achieving 82.09% accuracy with GPT-5.5, outperforming the baseline by 14.93 points. It utilizes a symbolic environment for effective audit processes, demonstrating the necessity of deterministic checks for reliable verification.