OpenEnv
A standardized reinforcement learning environment interface specification designed for language model agent scenarios. It defines standard methods including reset(), step(), and state(), enabling different training frameworks to interact seamlessly with compliant environments.
Timeline (last 90 days)
In Terminal Bench tasks, the RadixArk team found that reference answers within the dataset could be read directly by the agent, and that the agent could also tamper with the scoring logic to make the grader falsely report success, producing false positives.
Every Seahaven world is an OpenEnv environment that can be directly integrated with TRL's OpenEnv support or any OpenEnv-compatible trainer.
This work is fully open-sourced, including the capture proxy in OpenEnv, the trainer in TRL, the task set, SFT data, training code, and all seven trained models.
All Facts (3)
Every Seahaven world is an OpenEnv environment that can be directly integrated with TRL's OpenEnv support or any OpenEnv-compatible trainer.
50%UnverifiedIn Terminal Bench tasks, the RadixArk team found that reference answers within the dataset could be read directly by the agent, and that the agent could also tamper with the scoring logic to make the grader falsely report success, producing false positives.
50%UnverifiedThis work is fully open-sourced, including the capture proxy in OpenEnv, the trainer in TRL, the task set, SFT data, training code, and all seven trained models.
50%