This repo is based on the original Reflexion framwork Reflexion: Language Agents with Verbal Reinforcement Learning by Noah Shinn, Federico Cassano, Edward Berman, Ashwin Gopinath, Karthik Narasimhan, Shunyu Yao.

This project specifically targets reasoning tasks tailored for Arabic language by adapting self-reflecting agents.
The code now support arabic langugae processing and can be easily adapted with an API endpoint.

NOTE 1: The project is currently running on OpenAI API since the finetuned model is in a private repo.
NOTE 2: The instruction dataset currently is not publicly available.
NOTE 3: assign your API key in the agents_ar.py to try our project.

To run the project:

chainlit run main.py -w

• ReflexionStrategy.NONE - The agent is not given any information about its last attempt.

• ReflexionStrategy.LAST_ATTEMPT - The agent is given its reasoning trace from its last attempt on the question as context.

• ReflexionStrategy.REFLEXION - The agent is given its self-reflection on the last attempt as context.

• ReflexionStrategy.LAST_ATTEMPT_AND_REFLEXION - The agent is given both its reasoning trace and self-reflection on the last attempt as context.