KC-Agent: A Dual-Process Cognitive Architecture for Efficient ML Model Improvement

2026/08/03 by Gusseppe Bravo-Rocca, Jordi Guitart, Ajay Dholakia +2
Computer Science · #cs.AI

paper · pdf

Accepted at IEEE COMPSAC 2026

arxiv created 2026/08/03 · arxiv updated 2026/08/04

Abstract

Data drift poses significant challenges for machine learning systems in production, requiring continuous model updates to maintain performance. We present KC-Agent, a dual-process cognitive architecture for automated ML model improvement that combines fast pattern recognition (System 1) with deliberate incremental updates (System 2). Our approach implements structured memory systems enabling System 1 to leverage successful solutions previously discovered by System 2, achieving efficient pattern-based responses without costly re-computation. KC-Agent incorporates atomic change principles and rollback capabilities to ensure reliable, verifiable updates in production environments. We evaluate our method on five datasets including real-world NASA turbofan data with authentic temporal degradation and synthetic datasets with controlled drift scenarios. KC-Agent achieves state-of-the-art performance (76.8% accuracy) while maintaining optimal efficiency (13.2s execution time), outperforming established cognitive architectures: CodeAct (+2.4%), Tree of Thoughts (+3.6%), ReAct (+8.0%), and Reflexion (+8.9%). Consensus evaluation by a panel of state-of-the-art LLMs confirms superior strategic efficacy (8.33/10 Smartness score), significantly outperforming baseline agents. The knowledge consolidation mechanism delivers 91% speedup over the slow variant while maintaining higher accuracy. Our approach demonstrates both theoretical foundations and practical viability for cognitive-inspired automated ML improvement systems capable of handling complex real-world data drift scenarios.

Citations

• Towards a Science of Scaling Agent Systems
• Agents Thinking Fast and Slow: A Talker-Reasoner Architecture
• Self-Discover: Large Language Models Self-Compose Reasoning Structures
• Executable Code Actions Elicit Better LLM Agents
• Large Language Models Empowered Agent-based Modeling and Simulation: A Survey and Perspectives
• Monitoring Machine Learning Models: Online Detection of Relevant Deviations
• AgentBench: Evaluating LLMs as Agents
• Mind2Web: Towards a Generalist Agent for the Web
• SwiftSage: A Generative Agent with Fast and Slow Thinking for Complex Interactive Tasks
• Tree of Thoughts: Deliberate Problem Solving with Large Language Models
• Plan-and-Solve Prompting: Improving Zero-Shot Chain-of-Thought Reasoning by Large Language Models
• Generative Agents: Interactive Simulacra of Human Behavior
• Reflexion: Language Agents with Verbal Reinforcement Learning
• AI Maintenance: A Robustness Perspective
• A monitoring framework for deployed machine learning models with supply chain examples
• Distributing Accountability, Not Capability: Phase Separation and the LLM Workflow Quadrant in Autonomous AI Agent Architectures
• Evaluating Large Language Models Trained on Code
• Dark Experience for General Continual Learning: a Strong, Simple Baseline
• Don't forget, there is more than forgetting: new metrics for Continual\n Learning
• Learning to Learn without Forgetting by Maximizing Transfer and Minimizing Interference
• Detecting and Correcting for Label Shift with Black Box Predictors