analiza aiqbench
ReclAIm introduces an LLM-driven multi-agent framework that automates the detection and remediation of performance drift in medical imaging models, successfully restoring accuracy to within 2% of baseline after significant degradation. By integrating data augmentation and parameter-anchoring regularization, the system effectively mitigates catastrophic forgetting while managing class imbalance across diverse datasets like brain MRI and chest CT. This approach offers a practical pathway for maintaining robustness in clinical AI deployments without manual intervention.
ReclAIm: A Multi-Agent Framework for Monitoring and Correcting Performance Decline in Medical Imaging AI
arXiv:2510.17004v2 Announce Type: replace-cross Abstract: Purpose: To develop and evaluate a multi-agent framework (ReclAIm) for automated monitoring, detection, and correction of performance decline in medical image classification models. Materials and Methods: ReclAIm is a large language model-based multi-agent system that operates through natural language interaction. A master agent coordinating three task-specific agents performed performance evaluation and triggered fine-tuning when substantial performance declines were detected. The fine-tuning workflow incorporated data augmentation, class imbalance handling, and a parameter-anchoring regularization strategy to limit catastrophic forgetting. The system was benchmarked using multiple imaging datasets, including brain MRI, chest CT, and chest radiography, partitioned into model development, inference (performance monitoring), and fine-tuning subsets (60%:20%:20%). Results: ReclAIm successfully orchestrated training, evaluation, and performance monitoring across all datasets. Performance discrepancies between test and inference data were detected in 8 of 18 models, prompting fine-tuning workflows that reduced performance gaps. In cases with declines of up to 40.6% (cardiomegaly dataset, InceptionV3), fine-tuning restored performance metrics to within 2% of baseline values. Conclusion: ReclAIm provides a prototype framework for automated monitoring and targeted fine-tuning of medical image classification models, with a natural language interface designed to support accessibility in research and potential clinical applications.
Abstract:Purpose: To develop and evaluate a multi-agent framework (ReclAIm) for automated monitoring, detection, and correction of performance decline in medical image classification models. Materials and Methods: ReclAIm is a large language model-based multi-agent system that operates through natural language interaction. A master agent coordinating three task-specific agents performed performance evaluation and triggered fine-tuning when substantial performance declines were detected. The fine-tuning workflow incorporated data augmentation, class imbalance handling, and a parameter-anchoring regularization strategy to limit catastrophic forgetting. The system was benchmarked using multiple imaging datasets, including brain MRI, chest CT, and chest radiography, partitioned into model development, inference (performance monitoring), and fine-tuning subsets (60%:20%:20%). Results: ReclAIm successfully orchestrated training, evaluation, and performance monitoring across all datasets. Performance discrepancies between test and inference data were detected in 8 of 18 models, prompting fine-tuning workflows that reduced performance gaps. In cases with declines of up to 40.6% (cardiomegaly dataset, InceptionV3), fine-tuning restored performance metrics to within 2% of baseline values. Conclusion: ReclAIm provides a prototype framework for automated monitoring and targeted fine-tuning of medical image classification models, with a natural language interface designed to support accessibility in research and potential clinical applications.
Multiagent Systems (cs.MA); Artificial Intelligence (cs.AI)
arXiv:2510.17004 [cs.MA]
(or arXiv:2510.17004v2 [cs.MA] for this version)
https://doi.org/10.48550/arXiv.2510.17004 Focus to learn more arXiv-issued DOI via DataCite
https://doi.org/10.1148/ryai.250923 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 295 cuvinte
Subiecte conexe
RTSGameBench: An RTS Benchmark for Strategic Reasoning by Vision-Language Models
arXiv · 7 oct. 2026
Embodiment-Induced Coordination Regimes in Tabular Multi-Agent Q-Learning
arXiv · 28 iul. 2026
Learning Coordinated Preference for Multi-Objective Multi-Agent Reinforcement Learning
arXiv · 15 iun. 2026
OpenAgenet/OAN: Open Infrastructure for Trusted Agent Interconnection
arXiv · 4 iun. 2026