| name | student-experiment-design |
| description | Design experiments with variables, baselines, ablations, resources, and success criteria. |
Student Experiment Design
Student-specific extensions
- Add the exact artifact path where the plan will be published.
- Add
Reviewer-facing risks after ## Risks.
- The plan is incomplete if it cannot tell a reviewer what outcome would falsify the idea.
Purpose
Translate a research hypothesis into a concrete, executable experiment plan.
Workflow
- State hypothesis: What specific claim are you testing?
- Define variables: Independent (what you change), dependent (what you measure), controlled (what stays fixed)
- Choose baselines: What are you comparing against?
- Design protocol: Steps to execute, datasets, evaluation metrics
- Resource estimation: Compute, time, data requirements
- Publish: Write to your canonical shared artifact directory:
{artifact_dir}/{your-name}/experiment_plan_{id}.md
Output Format
# Experiment Plan: {title}
## Hypothesis
{Specific, testable claim}
## Variables
- **Independent**: What we vary (e.g., attention mechanism type)
- **Dependent**: What we measure (e.g., perplexity, latency)
- **Controlled**: What stays fixed (e.g., model size, dataset, training steps)
## Baselines
1. {Baseline method} — why it's relevant
2. ...
## Datasets
| Dataset | Size | Purpose | Source |
|---|---|---|---|
## Metrics
| Metric | Purpose | Expected Direction |
|---|---|---|
## Protocol
1. Step-by-step execution plan
2. ...
## Ablations
What variations to test:
1. ...
## Resource Estimate
- GPU hours: ...
- Storage: ...
- Expected runtime: ...
## Success Criteria
What results would support/refute the hypothesis.
## Risks
What could go wrong and mitigation strategies.
## Reviewer-facing risks
The objections a reviewer is most likely to raise and how the plan addresses them.
Rules
- Every experiment must have at least one baseline comparison
- Always include ablation studies to isolate the contribution
- Fix random seeds and log all hyperparameters
- Plan for at least 3 runs with different seeds for statistical significance