princeton-nlp/blindfold-textgame

[NAACL 2021] Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents

25
/ 100
Experimental

This project helps AI researchers evaluate how well their text-based game agents understand and navigate virtual environments without direct state information. It takes an existing text game and an agent design, and outputs performance metrics showing how effectively the agent can play the game, make decisions, and achieve goals when only given text descriptions. This is for AI researchers and developers focused on natural language understanding and reinforcement learning in interactive text-based scenarios.

No commits in the last 6 months.

Use this if you are an AI researcher developing and testing intelligent agents for text-based games and want to evaluate their performance when 'blindfolded' (i.e., without access to the game's internal state).

Not ideal if you are looking for a ready-to-play text game agent or a tool for general natural language processing tasks outside of interactive game environments.

AI game agents Reinforcement learning Natural language understanding Interactive fiction Computational linguistics
No License Stale 6m No Package No Dependents
Maintenance 0 / 25
Adoption 5 / 25
Maturity 8 / 25
Community 12 / 25

How are scores calculated?

Stars

11

Forks

2

Language

Python

License

Last pushed

May 31, 2021

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/nlp/princeton-nlp/blindfold-textgame"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.