SpellingBee

Description

SpellingBee is an environment for evaluating agents on word formation and vocabulary knowledge. This environment wraps the SpellingBee implementation from TextArena, a framework for text-based game environments.

Capabilities

• English vocabulary and word formation
• Strategic word selection with length constraints
• Turn-by-turn competitive gameplay
• Evaluation across small and large vocabulary variants

Compute Requirements

SpellingBee does not require a sandbox. It has minimal compute requirements.

License

MIT.

Tasks

There are two splits: train (450 tasks) and test (450 tasks). Each split contains 50 tasks across each of 9 variants:

• SpellingBee-v0
• SpellingBee-v0-large
• SpellingBee-v0-large-raw
• SpellingBee-v0-large-train
• SpellingBee-v0-raw
• SpellingBee-v0-small
• SpellingBee-v0-small-raw
• SpellingBee-v0-small-train
• SpellingBee-v0-train

Each task is seeded for reproducibility.

Reward Structure

This is a sparse reward environment. Rewards are mapped from TextArena's native range of {-1, 0, 1} to {0.0, 0.5, 1.0} via (raw + 1) / 2.

We do not use LLM graders for this environment; reward is determined programmatically.

Data

Game state is generated procedurally by the TextArena engine using seeded randomness. No external data files are required.

Tools

Agents are given a single tool:

• submit_word(word): Submit a word using the allowed letters. The word must be at least as long as the previous word.

Time Horizon

SpellingBee is a multi-turn environment.

Environment Difficulty

Medium - requires vocabulary knowledge and strategic word selection.

Other Environment Requirements

This environment requires an OpenAI API key (passed via secrets) to power the LLM opponent.

Safety

Agents in SpellingBee interact only with a word game and have no access to external systems, the internet, or sensitive data. The environment does not present safety risks.

Citations

@software{textarena2024,
  author    = {Guertler, Leon and Banting, Wilfried and Pignatelli, Eduardo},
  title     = {TextArena},
  year      = {2024},
  publisher = {GitHub},
  url       = {https://github.com/LeonGuertler/TextArena}
}