SimpleNegotiation

Description

SimpleNegotiation is an environment for evaluating agents on resource negotiation and strategic trading. This environment wraps the SimpleNegotiation implementation from TextArena, a framework for text-based game environments.

Capabilities

• Strategic resource trading and negotiation
• Multi-turn decision making with accept/deny options
• Competitive gameplay against LLM opponents
• Evaluation across multiple difficulty variants

Compute Requirements

SimpleNegotiation 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:

• SimpleNegotiation-v0
• SimpleNegotiation-v0-long
• SimpleNegotiation-v0-long-raw
• SimpleNegotiation-v0-long-train
• SimpleNegotiation-v0-raw
• SimpleNegotiation-v0-short
• SimpleNegotiation-v0-short-raw
• SimpleNegotiation-v0-short-train
• SimpleNegotiation-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 three tools:

• make_offer(offered, requested): Make a trade offer. Specify resources to offer and request (e.g. offered='3 Sheep, 2 Ore', requested='5 Brick').
• accept_offer(): Accept the current incoming trade offer.
• deny_offer(): Deny the current incoming trade offer.

Time Horizon

SimpleNegotiation is a multi-turn environment.

Environment Difficulty

Medium - requires strategic negotiation and resource valuation.

Other Environment Requirements

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

Safety

Agents in SimpleNegotiation interact only with a negotiation 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}
}