#AI Architecture

Articles (24)

• 🇺🇸 Neural Dynamics Self-Attention for Spiking Transformers — 23/03/2026 [USA]

  arXiv:2603.19290v1 Announce Type: cross Abstract: Integrating Spiking Neural Networks (SNNs) with Transformer architectures offers a promising pathway to balance energy efficiency and performance, pa...

  Related: #Neuromorphic Computing
• 🇺🇸 An Agentic Multi-Agent Architecture for Cybersecurity Risk Management — 23/03/2026 [USA]

  arXiv:2603.20131v1 Announce Type: cross Abstract: Getting a real cybersecurity risk assessment for a small organization is expensive -- a NIST CSF-aligned engagement runs $15,000 on the low end, take...

  Related: #Cybersecurity
• 🇺🇸 LeWorldModel: Stable End-to-End Joint-Embedding Predictive Architecture from Pixels — 23/03/2026 [USA]

  arXiv:2603.19312v1 Announce Type: cross Abstract: Joint Embedding Predictive Architectures (JEPAs) offer a compelling framework for learning world models in compact latent spaces, yet existing method...

  Related: #Predictive Modeling
• 🇺🇸 AS2 -- Attention-Based Soft Answer Sets: An End-to-End Differentiable Neuro-Soft-Symbolic Reasoning Architecture — 20/03/2026 [USA]

  arXiv:2603.18436v1 Announce Type: new Abstract: Neuro-symbolic artificial intelligence (AI) systems typically couple a neural perception module to a discrete symbolic solver through a non-differentia...

  Related: #Neuro-Symbolic AI
• 🇺🇸 Box Maze: A Process-Control Architecture for Reliable LLM Reasoning — 20/03/2026 [USA]

  arXiv:2603.19182v1 Announce Type: new Abstract: Large language models (LLMs) demonstrate strong generative capabilities but remain vulnerable to hallucination and unreliable reasoning under adversari...

  Related: #Reasoning Reliability
• 🇺🇸 InfoMamba: An Attention-Free Hybrid Mamba-Transformer Model — 20/03/2026 [USA]

  arXiv:2603.18031v1 Announce Type: cross Abstract: Balancing fine-grained local modeling with long-range dependency capture under computational constraints remains a central challenge in sequence mode...

  Related: #Deep Learning
• 🇺🇸 Engineering Verifiable Modularity in Transformers via Per-Layer Supervision — 20/03/2026 [USA]

  arXiv:2603.18029v1 Announce Type: cross Abstract: Transformers resist surgical control. Ablating an attention head identified as critical for capitalization produces minimal behavioral change because...

  Related: #Model Interpretability
• 🇺🇸 The Phasor Transformer: Resolving Attention Bottlenecks on the Unit Circle — 19/03/2026 [USA]

  arXiv:2603.17433v1 Announce Type: cross Abstract: Transformer models have redefined sequence learning, yet dot-product self-attention introduces a quadratic token-mixing bottleneck for long-context t...

  Related: #Computational Efficiency
• 🇺🇸 Transformers are Bayesian Networks — 19/03/2026 [USA]

  arXiv:2603.17063v1 Announce Type: new Abstract: Transformers are the dominant architecture in AI, yet why they work remains poorly understood. This paper offers a precise answer: a transformer is a B...

  Related: #Probabilistic Models
• 🇺🇸 IRAM-Omega-Q: A Computational Architecture for Uncertainty Regulation in Artificial Agents — 18/03/2026 [USA]

  arXiv:2603.16020v1 Announce Type: new Abstract: Artificial agents can achieve strong task performance while remaining opaque with respect to internal regulation, uncertainty management, and stability...

  Related: #Uncertainty Regulation
• 🇺🇸 Interpretable Context Methodology: Folder Structure as Agentic Architecture — 18/03/2026 [USA]

  arXiv:2603.16021v1 Announce Type: new Abstract: Current approaches to AI agent orchestration typically involve building multi-agent frameworks that manage context passing, memory, error handling, and...

  Related: #Interpretability
• 🇺🇸 StatePlane: A Cognitive State Plane for Long-Horizon AI Systems Under Bounded Context — 17/03/2026 [USA]

  arXiv:2603.13644v1 Announce Type: new Abstract: Large language models (LLMs) and small language models (SLMs) operate under strict context window and key-value (KV) cache constraints, fundamentally l...

  Related: #Cognitive Systems
• 🇺🇸 Spatio-Semantic Expert Routing Architecture with Mixture-of-Experts for Referring Image Segmentation — 16/03/2026 [USA]

  arXiv:2603.12538v1 Announce Type: cross Abstract: Referring image segmentation aims to produce a pixel-level mask for the image region described by a natural-language expression. Although pretrained ...

  Related: #Computer Vision
• 🇺🇸 Separable neural architectures as a primitive for unified predictive and generative intelligence — 13/03/2026 [USA]

  arXiv:2603.12244v1 Announce Type: cross Abstract: Intelligent systems across physics, language and perception often exhibit factorisable structure, yet are typically modelled by monolithic neural arc...

  Related: #Neural Networks
• 🇺🇸 DendroNN: Dendrocentric Neural Networks for Energy-Efficient Classification of Event-Based Data — 11/03/2026 [USA]

  arXiv:2603.09274v1 Announce Type: cross Abstract: Spatiotemporal information is at the core of diverse sensory processing and computational tasks. Feed-forward spiking neural networks can be used to ...

  Related: #Energy Efficiency
• 🇺🇸 Explainable Innovation Engine: Dual-Tree Agent-RAG with Methods-as-Nodes and Verifiable Write-Back — 11/03/2026 [USA]

  arXiv:2603.09192v1 Announce Type: new Abstract: Retrieval-augmented generation (RAG) improves factual grounding, yet most systems rely on flat chunk retrieval and provide limited control over multi-s...

  Related: #Explainable AI
• 🇺🇸 A Cortically Inspired Architecture for Modular Perceptual AI — 10/03/2026 [USA]

  arXiv:2603.07295v1 Announce Type: new Abstract: This paper bridges neuroscience and artificial intelligence to propose a cortically inspired blueprint for modular perceptual AI. While current monolit...

  Related: #Neuroscience
• 🇺🇸 Predictive Coding Graphs are a Superset of Feedforward Neural Networks — 09/03/2026 [USA]

  arXiv:2603.06142v1 Announce Type: cross Abstract: Predictive coding graphs (PCGs) are a recently introduced generalization to predictive coding networks, a neuroscience-inspired probabilistic latent ...

  Related: #Neural Networks
• 🇺🇸 The EpisTwin: A Knowledge Graph-Grounded Neuro-Symbolic Architecture for Personal AI — 09/03/2026 [USA]

  arXiv:2603.06290v1 Announce Type: new Abstract: Personal Artificial Intelligence is currently hindered by the fragmentation of user data across isolated silos. While Retrieval-Augmented Generation of...

  Related: #Personal AI
• 🇺🇸 RAGdb: A Zero-Dependency, Embeddable Architecture for Multimodal Retrieval-Augmented Generation on the Edge — 27/02/2026 [USA]

  arXiv:2602.22217v1 Announce Type: cross Abstract: Retrieval-Augmented Generation (RAG) has established itself as the standard paradigm for grounding Large Language Models (LLMs) in domain-specific, u...

  Related: #Edge Computing, #Privacy-Preserving Technology
• 🇺🇸 Characterizing State Space Model and Hybrid Language Model Performance with Long Context — 25/02/2026 [USA]

  arXiv:2507.12442v3 Announce Type: replace-cross Abstract: Emerging applications such as AR are driving demands for machine intelligence capable of processing continuous and/or long-context inputs on ...

  Related: #Hardware Optimization, #Long-Context Processing
• 🇺🇸 Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward — 16/02/2026 [USA]

  arXiv:2602.12430v1 Announce Type: cross Abstract: The transition from monolithic language models to modular, skill-equipped agents marks a defining shift in how large language models (LLMs) are deplo...

  Related: #Knowledge Management, #Model Efficiency
• 🇺🇸 Unrolling the Codex agent loop — 23/01/2026 [USA]

  A technical deep dive into the Codex agent loop, explaining how Codex CLI orchestrates models, tools, prompts, and performance using the Responses API.

  Related: #Technology, #Developer Tools
• 🇺🇸 How we built OWL, the new architecture behind our ChatGPT-based browser, Atlas — 30/10/2025 [USA]

  A deep dive into OWL, the new architecture powering ChatGPT Atlas—decoupling Chromium, enabling fast startup, rich UI, and agentic browsing with ChatGPT.

  Related: #Browser Innovation, #Human-AI Collaboration