Conceptual architecture
The references below describe ideas and proposed structure. They do not establish an implemented or evaluated system.
PROJECT / RESEARCH RELATIONSHIPS
Sanskrit Mandala Model
Architectural exposition and reading questions. Conceptual sources are kept separate from experimental findings.
RELATED PROJECTS
This page stays focused on Sanskrit Mandala Model. Use the project links to change projects, or the overview link to browse all projects. Counts refer to this selected set of extracts.
SMM Research
Selected SMM study extracts
These counts describe the curated selection, rather than the full research portfolio.
0 of 0 selected extracts
No study extracts selected for this view
This selection includes no study extract for this program. It does not describe the extent or outcome of its research.
KEEP THE DISTINCTIONS VISIBLE
Three different evidence questions
Implemented procedure
No selected study extract here establishes a procedure implemented around an existing model.
Trained-model work
No training or model-weight evidence is selected. A conceptual framework or a procedure does not establish model training.
READING & CONTEXT
SMM references
Claims from a reference remain scoped to that source.
SMM Layered AI Cognition Whitepaper
Enforcing invariant meaning preservation through concentric recursive layers.
Scope of this source
Read sourceAn architectural proposal. Experimental findings require separate study records.
SMM and UKM Distinction Public Thesis Brief
Establishing the complementary dynamics of structural processing and relational ontology.
Scope of this source
Read sourceA distinction between complementary frameworks and their responsibilities.
Structured Representation and Epistemic Integrity in AI Reasoning
A controlled prompt-scaffold ablation separates evaluator-rated epistemic integrity from solution quality and novelty.
Scope of this source
Read sourceAdjacent research. Its conclusions remain within the tasks and conditions reported in that paper.
Capability Is Not Its Encoding
A controlled Machine Edition experiment separates the demonstrated value of an explicit capability from the unproven incremental value of its machine-native encoding and deterministic runtime.
Scope of this source
Read sourceA conceptual boundary between a representation and the capability of an implemented system.
Sanskrit Mandala Model
A layered reference architecture for intelligence systems that need interpretability, bounded expansion, and alignment without flattening meaning.
Scope of this source
Read sourceThe published conceptual framework and its structural vocabulary.
The Sanskrit Mandala Model
A Layered Architecture for Interpretable and Aligned AI
Scope of this source
Read sourceBook-level exposition. Empirical results require separate study records.
FURTHER INQUIRY
Open questions
How can meaning remain inspectable through change?
A future evaluation would need a defined task, explicit transformations, and checks that distinguish meaning preservation from fluent output.
Where do SMM and UKM keep distinct responsibilities?
A comparison should make each framework's role and scope explicit before attributing an outcome to their combination.
What evidence would support an architectural claim?
A proposed structure, an implemented capability, and an observed result each require their own account of what was examined.