9.2 KiB
Theoretical Framework: Expert-Augmented LLM Ideation
The Core Problem: LLM "Semantic Gravity"
What is Semantic Gravity?
When LLMs generate creative ideas directly, they exhibit a phenomenon we term "semantic gravity" - the tendency to generate outputs that cluster around high-probability regions of their training distribution.
Direct LLM Generation:
Input: "Generate creative ideas for a chair"
LLM Process:
P(idea | "chair") → samples from training distribution
Result:
- "Ergonomic office chair" (high probability)
- "Foldable portable chair" (high probability)
- "Eco-friendly bamboo chair" (moderate probability)
Problem:
→ Ideas cluster in predictable semantic neighborhoods
→ Limited exploration of distant conceptual spaces
→ "Creative" outputs are interpolations, not extrapolations
Why Does This Happen?
- Statistical Pattern Learning: LLMs learn co-occurrence patterns from training data
- Mode Collapse: When asked to be "creative," LLMs sample from the distribution of "creative ideas" they've seen
- Relevance Trap: Strong associations dominate weak ones (chair→furniture >> chair→marine biology)
- Prototype Bias: Outputs gravitate toward category prototypes, not edge cases
The Solution: Expert Perspective Transformation
Theoretical Basis
Our approach draws from three key theoretical foundations:
1. Semantic Distance Theory (Mednick, 1962)
"Creative thinking involves connecting weakly related, remote concepts in semantic memory."
Key insight: Creativity correlates with semantic distance. The farther the conceptual "jump," the more creative the result.
Our application: Expert perspectives force semantic jumps that LLMs wouldn't naturally make.
Without Expert:
"Chair" → furniture, sitting, comfort, design
Semantic distance: SHORT
With Marine Biologist Expert:
"Chair" → underwater pressure, coral structure, buoyancy, bioluminescence
Semantic distance: LONG
Result: Novel ideas like "pressure-adaptive seating" or "coral-inspired structural support"
2. Conceptual Blending Theory (Fauconnier & Turner, 2002)
"Creative products emerge from blending elements of two input spaces into a novel integrated space."
The blending process:
- Input Space 1: The target concept (e.g., "chair")
- Input Space 2: The expert's domain knowledge (e.g., marine biology)
- Generic Space: Abstract structure shared by both
- Blended Space: Novel integration of elements from both inputs
Our application: Each expert provides a distinct input space for systematic blending.
┌─────────────────┐ ┌─────────────────┐
│ Input 1 │ │ Input 2 │
│ "Chair" │ │ Marine Biology │
│ - support │ │ - pressure │
│ - sitting │ │ - buoyancy │
│ - comfort │ │ - adaptation │
└────────┬────────┘ └────────┬────────┘
│ │
└───────────┬───────────┘
▼
┌─────────────────────┐
│ Blended Space │
│ Novel Chair Ideas │
│ - pressure-adapt │
│ - buoyant support │
│ - bio-adaptive │
└─────────────────────┘
3. Design Fixation Breaking (Jansson & Smith, 1991)
"Design fixation is blind adherence to initial ideas, limiting creative output."
Fixation occurs because:
- Knowledge is organized around category prototypes
- Prototypes require less cognitive effort to access
- Initial examples anchor subsequent ideation
Our application: Expert perspectives act as "defixation triggers" by activating non-prototype knowledge.
Without Intervention:
Prototype: "standard four-legged chair"
Fixation: Variations on four-legged design
With Expert Intervention:
Archaeologist: "Ancient people sat differently..."
Dance Therapist: "Seating affects movement expression..."
Fixation Broken: Entirely new seating paradigms explored
The Multi-Expert Aggregation Model
From "Wisdom of Crowds" to "Inner Crowd"
Research shows that groups generate more diverse ideas because each member brings different perspectives. Our system simulates this "crowd wisdom" through multiple expert personas:
Traditional Crowd:
Person 1 → Ideas from perspective 1
Person 2 → Ideas from perspective 2
Person 3 → Ideas from perspective 3
Aggregation → Diverse idea pool
Our "Inner Crowd":
LLM + Expert 1 Persona → Ideas from perspective 1
LLM + Expert 2 Persona → Ideas from perspective 2
LLM + Expert 3 Persona → Ideas from perspective 3
Aggregation → Diverse idea pool (simulated crowd)
Why Multiple Experts Work
- Coverage: Different experts activate different semantic regions
- Redundancy Reduction: Deduplication removes overlapping ideas
- Diversity by Design: Expert selection can be optimized for maximum diversity
- Diminishing Returns: Beyond ~4-6 experts, marginal diversity gains decrease
The Complete Pipeline
Stage 1: Attribute Decomposition
Purpose: Structure the problem space before creative exploration
Input: "Innovative chair design"
Output:
Categories: [Material, Function, Usage, User Group]
Material: [wood, metal, fabric, composite]
Function: [support, comfort, mobility, storage]
Usage: [office, home, outdoor, medical]
User Group: [children, elderly, professionals, athletes]
Theoretical basis: Structured decomposition prevents premature fixation on holistic solutions.
Stage 2: Expert Team Generation
Purpose: Assemble diverse perspectives for maximum semantic coverage
Strategies:
1. LLM-Generated: Query-specific, prioritizes unconventional experts
2. Curated: Pre-selected high-quality occupations
3. External Sources: DBpedia, Wikidata for broad coverage
Diversity Optimization:
- Domain spread (arts, science, trades, services)
- Expertise level variation
- Cultural/geographic diversity
Stage 3: Expert Transformation
Purpose: Apply each expert's perspective to each attribute
For each (attribute, expert) pair:
Input: "Chair comfort" + "Marine Biologist"
LLM Prompt:
"As a marine biologist, how might you reimagine
chair comfort using principles from your field?"
Output: Keywords + Descriptions
- "Pressure-distributed seating inspired by deep-sea fish"
- "Buoyancy-assisted support reducing pressure points"
Stage 4: Deduplication
Purpose: Ensure idea set is truly diverse, not just numerous
Methods:
1. Embedding-based: Fast cosine similarity clustering
2. LLM-based: Semantic pairwise comparison (more accurate)
Output:
- Unique ideas grouped by similarity
- Representative idea selected from each cluster
- Diversity metrics computed
Stage 5: Novelty Validation
Purpose: Ground novelty in real-world uniqueness
Process:
- Search patent databases for similar concepts
- Compute overlap scores
- Flag ideas with high existing coverage
Output:
- Novelty score per idea
- Patent overlap rate for idea set
Testable Hypotheses
H1: Semantic Diversity
Multi-expert generation produces higher semantic diversity than single-expert or direct generation.
Measurement: Mean pairwise cosine distance between idea embeddings
H2: Novelty
Ideas from multi-expert generation have lower patent overlap than direct generation.
Measurement: Percentage of ideas with existing patent matches
H3: Expert Count Effect
Semantic diversity increases with expert count, with diminishing returns beyond 4-6 experts.
Measurement: Diversity vs. expert count curve
H4: Expert Source Effect
LLM-generated experts produce more unconventional ideas than curated/database experts.
Measurement: Semantic distance from query centroid
H5: Fixation Breaking
Multi-expert approach produces more ideas outside the top-3 semantic clusters than direct generation.
Measurement: Cluster distribution analysis
Expected Contributions
- Theoretical: Formalization of "semantic gravity" as LLM creativity limitation
- Methodological: Expert-augmented ideation pipeline with evaluation framework
- Empirical: Quantitative evidence for multi-expert creativity enhancement
- Practical: Open-source system for innovation ideation
Positioning Against Related Work
| Approach | Limitation | Our Advantage |
|---|---|---|
| Direct LLM generation | Semantic gravity, fixation | Expert-forced semantic jumps |
| Human brainstorming | Cognitive fatigue, social dynamics | Tireless LLM generation |
| PersonaFlow (2024) | Research-focused, no attribute structure | Product innovation, structured decomposition |
| PopBlends (2023) | Two-concept blending only | Multi-expert, multi-attribute blending |
| BILLY (2025) | Vector fusion less interpretable | Sequential generation, explicit control |