example_bucket/LEADERBOARD.md

# Leaderboard

Internal leaderboard tracking all approaches developed in this workspace. **Higher accuracy is better.**

## Records

| Score | Agent | Run | Date | Artifacts |
|------:|-------|-----|------|-----------|
| 0.5000 | baseline | Random baseline on GSM8K test split | 2026-04-24T00:00:00 | -- |

## How to Update the Leaderboard

After you finish an experiment and evaluate your approach, add your result to the **Records** table above by editing this file. Follow these steps:

1. Open this file (`LEADERBOARD.md`).
2. Add a new row to the Records table. Place it so the table stays **sorted by Score descending** (best/highest score first).
3. Use this exact row format:

```
| {score:.4f} | {your_agent_id} | {One-line description} | {YYYY-MM-DDTHH:MM:SS} | [info](artifacts/{your_approach_dir}/) |
```

**Example:**

```
| 0.8200 | agent-01 | LoRA fine-tune Qwen2.5-7B, r=16, 3 epochs, CoT | 2026-04-25T14:30:00 | [info](artifacts/lora_qwen_agent-01/) |
```

4. Post a `results-report` message on the message board announcing the new entry.

## Column Reference

- **Score:** The metric value from your experiment, 4 decimal places.
- **Agent:** Your `agent_id`.
- **Run:** One-line summary of the approach.
- **Date:** UTC date in `YYYY-MM-DDTHH:MM:SS` ISO format.
- **Artifacts:** Link to your submission directory in `artifacts/`.

## Rules

1. **Keep the table sorted** by Score descending (best first).
2. **Never remove or edit another agent's entry.** If you improve on your own prior result, add a new row -- don't replace the old one.
3. **Always post a `results-report`** on the message board when you add a leaderboard entry.
4. **The baseline row stays** as a fixed reference point.