Cognitive Architectures for AI Agents

Overview

Cognitive architectures study how to organize an agent’s perception, memory, goals, reasoning, and actions so it can behave coherently over time. Approaches range from symbolic and BDI-style systems, state machines, and global-workspace models to LLM-based agents with memory, tools, reflection, and specialized components; many combine these methods with learning or embodied control. Researchers evaluate them on long-horizon tasks such as navigation, household and mobile-device control, tool use, multi-agent coordination, and robot deployment, while also examining safety and human oversight. A central challenge is making agents persistent, grounded, and reliable without sacrificing adaptability, efficiency, or interpretability; adding cognitive modules alone does not guarantee better performance.

Latest papers

92 papers6 papers in the last four weeks, down 25% on the four weeks before. 0.0% of all new papers.

Feb 18, 2026

HiVAE: Hierarchical Latent Variables for Scalable Theory of Mind

Theory of mind (ToM) enables AI systems to infer agents' hidden goals and mental states, but existing approaches focus mainly on small human understandable gridworld spaces. We introduce HiVAE, a hierarchical variational architecture that scales ToM reasoning to realistic spatiotemporal domains. Inspired by the belief-desire-intention structure of human cognition, our three-level VAE hierarchy achieves substantial performance improvements on a 3,185-node campus navigation task. However, we identify a critical limitation: while our hierarchical structure improves prediction, learned latent representations lack explicit grounding to actual mental states. We propose self-supervised alignment strategies and present this work to solicit community feedback on grounding approaches.
Sep 23, 2025

Structured Cognitive Loop for Behavioral Intelligence in Large Language Model Agents (Extended Revision: From Behavioral Architecture to Epistemic Accountability)

The central challenge for AI agents is not performance but accountability. An agent that produces correct outputs through an opaque process offers no basis for verifying why a decision was permitted, attributing errors to identifiable causes, or delegating authority with justified confidence. We introduce the Structured Cognitive Loop (SCL), an architecture that separates cognition, control, action, and memory: the language model proposes, a lightweight controller mediates every proposal before action, and memory is kept externally as a structured store of verified facts. Against ReAct and common LangChain agents on three tasks under matched models, tools, and decoding, SCL achieves 86.3% task success versus 70.5-76.8% across 360 episodes, with higher goal fidelity, fewer redundant calls, more reliable reuse of intermediate states, and fewer unsupported assertions; ablations show that control and external memory contribute independently. We then situate SCL within a broader accountability architecture comprising context-aware human-in-the-loop control, Pool-Gated Retrieval, and the Horizon-Warrant-Commitment framework, and outline the current direction of development, in which instructions are compiled into programs validated before execution so that execution authority derives from a verified program rather than per-cycle model interpretation. For the kind of accountability it pursues, SCL adopts neither multi-agent consensus as a proxy for truth nor the imitation of human cognition as a design goal: it institutionalizes cognition rather than imitating it.