Symbolic Prompting - Variables (For Educationl Use Only)
🏠 Home
There exist different way to define variables as part of Symbolic Prompt.
Use Atomic, Unambiguous Names
GOOD BAD essay_difficulty_level feedback_required student_grade_level level Thing status grade
Apply Consistent Naming Conventions
CONVENTION
PURPOSE
EXAMPLE
snake_case
Mutable variables
essay_difficulty_level
UPPER_SNAKE_CASE
Constants / ENUM types
DIFFICULTY_LEVEL
PascalCase
Types / schemas
EssayType
lowercase values
Free text (limited)
"literary analysis"
UPPERCASE values
ENUM members
BEGINNER, INTERMEDIATE, ADVANCED
DEFINITION INTERPRETER EXAMPLE TYPE_ESSAY = Constant / Enum Global / Value not change TYPE_ESSAY = NARRATIVE | EXPOSITORY | ARGUMENTATIVE type_essay = Variable or Constant / Update value type_essay = {narrative, expository, argumentative, descriptive} Type_Essay = {... Type or Class Type_Essay = Narrative | Expository | Argumentative TypeEssay = {... Type or Class TypeEssay = { category: EssayCategory, word_count: Integer, required_sections: List } Compact Syntax Not Recommended typeessay={lit,pers,exp,desc}
Variable Definition Method Comparison
METHOD
SYNTAX
BEST FOR
EXAMPLE
LLM CLARITY
Explicit Typing
variable: Type = value
Formal systems, type safety
difficulty: DifficultyLevel = Intermediate
High
Set Membership
variable ∈ {values}
Discrete choices, enumerations
essay_type ∈ {Narrative, Expository, Persuasive}
High
Mathematical Range
variable ∈ [min..max]
Numerical ranges, intervals
min_pass_score ∈ [70..80]
High
Assignment
variable = value
Simple values, constants
min_pass_score = 70
Medium-High
Definition
variable := expression
Derived values, aliases
passing := student_score ≥ min_pass_score
Medium
Predicate Form
is_property(variable)
Boolean conditions, flags
is_advanced(essay)
Medium
Structured
variable = {field: value}
Objects, records, entities
essay = {id: "E123", type: Persuasive, score: 85}
High
Variable Definition Patterns by Use Case
USE CASE
RECOMMENDED PATTER
EXAMPLE
WHY IT WORKS
Essay Classification
Enumeration with membership
difficulty_level ∈ {Beginner, Intermediate, Advanced, Expert}
Clear categories, easy validation
Lesson Planning
Structured objects
lesson_plan = {topic: String, duration: Minutes, objectives: List<String>, materials: List<String>}
Captures complex relationships
Student Assessment
Derived booleans
is_proficient := (score ≥ 80) ∧ (completion_rate ≥ 0.9)
Computable conditions
Assignment Management
Typed variables with ranges
due_date: Date ∈ [today..today+7]
Quantitative constraints
Curriculum Design
Set operations
advanced_topics ⊆ all_topics ∩ (math_topics ∪ science_topics)
Mathematical precision
Essay Evaluation
Predicate functions
has_thesis_statement(essay)
Functional approach
Symbol Selection for Variable Types
VARIABLE TYPE
RECOMMENDED SYMBOL
EXAMPLE
ALTERNATIVE
Enumeration
∈ {values}
essay_type ∈ {Narrative, Expository}
: EnumType
Range
∈ [min..max]
grade_level ∈ [1..12]
: NumberRange
Boolean
: Boolean
has_conclusion: Boolean
∈ {true, false}
String
: String
student_name: String
= "value"
Integer
: Integer
page_count: Integer
∈ ℤ
Float
: Float
essay_score: Float
∈ ℝ
List
: List<Type>
topics: List<String>
= [value1, value2]
Set
: Set<Type>
prerequisites: Set<Course>
⊆ {values}
Object
: ObjectType
submission: StudentWork
= {field: value}
GOOD BAD DIFFICULTY_LEVEL = {BEGINNER, INTERMEDIATE, ADVANCED, EXPERT} essay_difficulty in DIFFICULTY_LEVEL d = {b, i, a, e} essay_difficulty = advanced
Prefer Closed Sets (ENUM) Over free Text
GOOD
BAD
feedback_type in {GRAMMAR, STRUCTURE, CONTENT, CITATIONS}
feedback_type = "whatever the essay needs"
GOOD
BAD
needs_revision = true
revision_status = "pending"
has_thesis = false
thesis_present = "no"
is_plagiarized = false
plagiarism_check = "clean"
Avoid Reassigning Constants
Constants define ontology, not state.
GOOD BAD DIFFICULTY_LEVEL = {BEGINNER, INTERMEDIATE, ADVANCED, EXPERT} essay_difficulty = ADVANCED DIFFICULTY_LEVEL = ADVANCED
⚖️ Legal Disclaimer (Click to expand)
This repository is for educational purposes only regarding Symbolic Prompting. The author is not responsible for the use that third parties may make of these techniques. The user is responsible for respecting the terms of service of AI platforms and applicable legislation. All content is provided "AS IS," without warranties.
Compatibility may vary depending on model updates, tokenization behavior, and symbol parsing.
⭐ If this class helped you think differently about LLMs, consider starring the repository.
🏠 Home