Dr Ran Liu · UniBridgeAI
sophieranliu.bsky.social
Dr Ran Liu · UniBridgeAI
@sophieranliu.bsky.social
AI researcher & UniBridgeAI founder.
I build AI knowledge frameworks and cross-domain methods to help people learn AI more effectively and apply it in their own domains. More: https://zenodo.org/communities/unibridgeai-frameworks
Pinned
AI is moving fast & fragmented. This Cognitive Hologram makes it navigable: manageable for learning, useful for planning, transferable across domains. It's top-level map, future posts are zoom-ins of its components.
www.youtube.com/watch?v=Ib5c...
DOI: doi.org/10.5281/zeno...
#AI #MachineLearning
A New Way to Learn AI That Scales: The Cognitive Hologram of AI Knowledge
YouTube video by UniBridgeAI
www.youtube.com
AI will transform education—but how? This AI for Education Domain Compass shows where AI reshapes education (learner formation, learning practice, ecosystem) and how that happens through the underlying mechanisms—core AI functions.
doi.org/10.5281/zeno...
#AIEducation #AcademicSky #MachineLearning
AI for Education Domain Compass: A Strategic Map for AI Integration in Education
The AI for Education Domain Compass is a decision framework for making AI-in-education work legible, comparable, and designable at the system level. Instead of treating “AI in education” as a collecti...
doi.org
December 23, 2025 at 6:54 PM
AI is moving fast & fragmented. This Cognitive Hologram makes it navigable: manageable for learning, useful for planning, transferable across domains. It's top-level map, future posts are zoom-ins of its components.
www.youtube.com/watch?v=Ib5c...
DOI: doi.org/10.5281/zeno...
#AI #MachineLearning
A New Way to Learn AI That Scales: The Cognitive Hologram of AI Knowledge
YouTube video by UniBridgeAI
www.youtube.com
December 12, 2025 at 10:10 PM
Introducing AI CityMap:when a model behaves wrong or you want a specific effect, it tells you where to look and where to intervene, so your fixes become targeted instead of trial-and-error. www.youtube.com/watch?v=q-5E...
DOI: doi.org/10.5281/zeno...
#AI #MachineLearning #AIFramework #AIKnowledge
The AI City Map — State-Based AI Backbone
YouTube video by UniBridgeAI
www.youtube.com
December 12, 2025 at 7:55 PM
A New Way to Understand AI: a capability-level map that stays stable even as models and buzzwords keep changing. If you want a clearer way to plan, compare, and apply AI, this is for you! www.youtube.com/watch?v=X3E7...
DOI: doi.org/10.5281/zeno...
#AI #MachineLearning #AIFramework #AIKnowledge
A New Way to Understand AI: The 6 Core Functions
YouTube video by UniBridgeAI
www.youtube.com
December 12, 2025 at 7:52 PM
If deep learning workflows still feel messy, this short video walks through the Process-based AI Backbone (6 modules, 3 loops) and shows how to use it to debug models and plan AI-for-X projects. www.youtube.com/watch?v=Poj_...
#AI #MachineLearning #AcademicSky
The Process-based AI Backbone (Deep Learning Process)
YouTube video by UniBridgeAI
www.youtube.com
December 11, 2025 at 8:28 PM
A simple LLM email-assistant example to show how the process- & state-based AI backbones can be used in practice: the diagram illustrates how they structure the pipeline and where to look when behavior goes wrong.
#AI #MachineLearning #AcademicSky
December 11, 2025 at 3:16 PM
– process view of AI: see any deep learning workflow (doi.org/10.5281/zeno...)
– state view of AI: see models as six neural state spaces
(doi.org/10.5281/zeno...)

Together they give a stable skeleton for reasoning about any complex AI systems. #AI #MachineLearning #AcademicSky
December 11, 2025 at 2:27 AM
State-based AI backbone is the second zoom-in under the Cognitive Hologram: it shows that any deep learning system can be viewed as six spaces, giving a stable spatial lens to locate behaviors, diagnose failures, and decide where to intervene. #AI #MachineLearning doi.org/10.5281/zeno...
State-Based AI Backbone (Neural State Spaces)
This figure expands the state‑based view of the Foundational AI Backbone from the “Cognitive Hologram of AI Knowledge” (DOI: 10.5281/zenodo.17860776). While the process‑based backbone (DOI: 10.5281/ze...
doi.org
December 10, 2025 at 9:09 PM
Process-based AI backbone is the first zoom-in under the Cognitive Hologram: it shows that any deep learning workflow can be organized into six core modules & three loops (training, evaluation, adaptation), provides a stable backbone for understanding & designing AI. #AI doi.org/10.5281/zeno...
Process‑Based AI Backbone (Deep Learning Process)
This figure expands the process‑based view of the Foundational AI Backbone from the “Cognitive Hologram of AI Knowledge” (DOI: 10.5281/zenodo.17860776). While the high‑level Cognitive Hologram shows h...
doi.org
December 10, 2025 at 9:06 PM
Here's a compact map of AI knowledge that shows, in one picture, how data flows, what’s happening inside the deep learning network, and where AI plugs into your own field.

For more information, visit: doi.org/10.5281/zeno...

#AI #MachineLearning #AIFramework #AIKnowledge
The Cognitive Hologram of AI Knowledge
The Cognitive Hologram of AI Knowledge is a transdimensional learning architecture that organizes AI knowledge as a three‑layer cognitive structure. At its core lies a Transdimensional Learning Engine...
doi.org
December 8, 2025 at 10:57 PM
Think AI is just a smart assistant?
The Four-Tier Framework shows how much more AI can actually do with you and for you – far beyond what you can imagine. For more information, visit: doi.org/10.5281/zeno...

#AI #MachineLearning #AIFramework
December 8, 2025 at 7:00 PM
If AI feels like a blur of models and buzzwords, this map shows the six core functions it all reduces to.

To see how this simple backbone supports different disciplines in more detail, visit zenodo.org/communities/...

#AI #MachineLearning #AIFramework #AICapabilities #AIResearch
December 6, 2025 at 4:38 PM
🎨 AI for Art Domain Compass: a map of how AI can shape meaning, form, and context in artworks, helping artists and researchers make sense of existing AI-art practices and design new ones more intentionally.
doi.org/10.5281/zeno...
#AIArt #CreativeAI #AcademicSky #MachineLearning
AI for Art Domain Compass: A Structural Map for Designing AI's Role in Art
Unlike conventional taxonomies or tool lists, this Compass offers a holistic view of the AI-art space, provides a compact and stable skeleton for understanding a complex and rapidly evolving landscape...
doi.org
December 5, 2025 at 10:30 PM
🔬 AI for Science Strategic Compass
A strategic decision framework that helps scientists design well-informed AI strategies aligned with their scientific discovery challenges.
doi.org/10.5281/zeno...
#AIForScience #AcademicSky #AI #MachineLearning #ML4Science
AI for Science Strategic Compass (AFSC) – Full Strategy Matrix
AFSC Strategy Matrix The AI for Science Strategic Compass (AFSC) is a function-based decision framework that helps scientists make strategy-level choices about how to use AI to address specific scient...
doi.org
December 5, 2025 at 8:21 PM
AI for Marketing Domain Compass: a simple map that turns scattered AI tools into a clear picture of AI marketing. It helps you see where any AI project sits, what core capabilities it uses, & which areas are crowded or still open to explore. doi.org/10.5281/zeno...
#AIForMarketing #academicsky
AI for Marketing Domain Compass: Mapping and Guiding AI-Enabled Marketing
The AI for Marketing Domain Compass is a structural framework that organizes AI-enabled marketing into a coherent “AI Marketing Space.” It provides a compact, system-level view of how AI intervenes in...
doi.org
December 5, 2025 at 7:42 PM
🧱 Four Integration Tiers for Human–AI Collaboration
A simple map of how deeply AI is woven into a workflow: from on-demand assistants to embedded engines, domain-adapted systems, and AI-native operating layers. It helps you see where you are now and where you want to go. doi.org/10.5281/zeno...
Four-Tier Framework for Human-AI Collaboration
This work introduces a four-tier framework for Human–AI collaboration that takes a human- and organization-centric view. Instead of ranking AI systems by how “intelligent” or autonomous they are, the ...
doi.org
December 5, 2025 at 6:58 PM
🧩 AI Core Functions
A six-function map of AI’s core capabilities that lets you see the whole fast-evolving AI landscape, decompose any model or method into basic abilities, and design systems by choosing functions that directly align with the problem you’re solving.
DOI: doi.org/10.5281/zeno...
AI Core Function Ontology: A Two-Level Capability Framework
Existing AI taxonomies often organize methods by models, algorithms, or learning paradigms. This tends to produce a fragmented landscape in which the link between user intent and AI capability remains...
doi.org
December 5, 2025 at 6:50 PM
AI is changing fast and AI knowledge is scattered. Many of us feel we must learn and use AI, but chasing tools and models is exhausting and never feels enough.

I started UniBridgeAI to develop more structured and effective ways to learn AI and apply it in different domains.
December 5, 2025 at 6:05 PM