MENTOR
A multimodal AI teaching system for reflective architectural learning.
MENTOR is an AI-powered educational system designed to support deeper cognitive engagement in architectural education. Rather than functioning as a question-answering assistant, MENTOR guides students through structured reasoning, dialogue, and visual interpretation to encourage reflection, discovery, and independent thinking.
Developed and evaluated over nine weeks, the system integrates large language models, vision-language models, and agent-based orchestration to scaffold learning across conceptual, spatial, and technical dimensions. MENTOR was benchmarked against a GPT-only baseline, with results demonstrating improved engagement, reduced cognitive offloading, and increased depth of design reasoning.
The Story
2
5

5


MENTOR is built as a coordinated multi-agent system, where each agent supports a distinct pedagogical role.
System Architecture
Core Agents
1
Orchestration
2
Context Reasoning Agent
Interprets user input and situational context, acting as the entry point for all interactions.Knowledge Synthesis Agent
Connects ideas across domains, synthesizing theory, precedent, and technical knowledge.Socratic Dialogue Agent
Drives questioning, reflection, and conceptual exploration through guided prompts.Cognitive Enhancement Agent
Regulates cognitive load, pacing, and depth of engagement.
Knowledge Infrastructure
3
The workflows are implemented in LangGraph.
The Context Reasoning Agent routes interactions through single-agent or multi-agent coordination paths, including sequential, parallel, and adaptive flows. A synthesizer stage merges agent outputs into a coherent educational response.
Models
Microsoft Magma for vision-language reasoning
GPT-4 Turbo and Claude 3.5 Sonnet for layered interpretation and dialogue
MENTOR’s Domain Expert relies on a hybrid Retrieval-Augmented Generation pipeline.
RAG Pipeline
Persistent vector store using ChromaDB
Hybrid search combining semantic, keyword, and query expansion strategies
Smart chunking and metadata enrichment
Sentence-transformer embeddings
Educational Safeguards
Pedagogical filtering reformats retrieved content for learning contexts
Progressive disclosure reduces extraneous cognitive load
Citation tracking returns sources in APA, MLA, or Chicago style
Results are deduplicated and scored based on authority, relevance, and pedagogical suitability
Cognitive Enhancement
Analysis
Agent
Context
Agent
Domain
Expert




Socratic
Tutor

Knowledge
Transfer
Cognitive
Development
Performance
Optimization
Capability
Building
The Oracle
The Catalyst
The Core Transaction
The Outcome Design
The Interaction Dynamic
PDF Repository
local
Semantic Search
vector similarity
Context Analysis
user intent
Text Extraction
PyMuPDF
Keyword Matching
text matching
Response
multi-source
Text Cleaning
Preprocessing
Query Expansion
enhanced terms
Citation
tracking
Smart Chunking
Overlaps
Hybrid Search
merged result
Quality
format text
DOCUMENT INGESTION
SEARCH
STRATEGIES
CHROMA DB
KNOWLEDGE
SYNTHESIS
document
chunks
extraction
context
data
search
result
citation
enhanced
response
cleaned txt
vector query
hybrid
results
sentence
transformers
+
embeddings
semantic
search
keyword
match
centralized
state
knowledge
search
processor
knowledge
base
context
analysis
processor
knowledge
synthesis
processor

5


Learning Framework
4
Interactivity
5
The system is structured around three learning phases, each with distinct objectives and agent roles.
Phase 1: Ideation
Focuses on creativity, conceptual exploration, and questioning.
- Led by the Socratic Dialogue Agent, with minimal technical constraints.
Phase 2: Visualization
Bridges ideas with spatial representation.
- Balances creative inquiry and technical input using multimodal interaction and analysis support.
Phase 3: Materialization
Emphasizes technical knowledge, system integration, and professional practice.
- Guided primarily by the Domain Expert Agent and real-world constraints.
A weighted completion formula values questioning, conceptual clarity, and response quality, encouraging balanced progression across phases.
Task generation
MENTOR generates adaptive challenges aligned with each learning phase:
Open-ended exploration in ideation
Spatial reasoning and representation in visualization
Real-world technical constraints in materialization
Image Generation
Visual outputs translate abstract reasoning into diagrams, spatial layouts, and performance visualizations, reinforcing connections between concept and form.
Gamification
Progress metrics, phase completion formulas, and challenge-based learning maintain engagement. Learners advance through levels tied to ideation, visualization, and materialization, reinforcing both creativity and rigor.




5


Evaluation and Benchmarking
Cognitive Analysis
Student interactions are logged and transformed into interaction graphs.
Metrics include:
Cognitive offloading prevention
Depth of thinking
Design move connectivity
Automated fuzzy linkography identifies critical moves and knowledge gaps.
Machine Learning Integration
Graph Neural Networks detect higher-order cognitive patterns
Ensemble classifiers assess proficiency levels
Statistical validation using mixed-effects models, ANOVAs, and Bayesian hierarchical analysis
Dashboards visualize learning progress, agent performance, and cognitive metrics in real time.
Results
MENTOR is designed to help students think through design decisions, not around them.
MENTOR demonstrated measurable improvements over a GPT-only baseline:
84.7% cognitive offloading prevention rate
73.9% deeper thinking engagement rate
Students spent more time reasoning through decisions, articulating intent, and reflecting on alternatives rather than accepting generated answers.
MENTOR shows how multimodal, agent-based AI systems can function as cognitive scaffolds rather than content generators. By structuring dialogue, vision, and knowledge retrieval around pedagogical goals, the system supports reflective, discovery-based learning in architectural education.
Project developed in collaboration with Seda Soylu and Biel Pitman. You can view the original article here, and the published paper here.
