---
license: mit
base_model:
- Qwen/Qwen2.5-3B-Instruct
---
This is the baseline checkpoint for paper: [**ToMAP: Training Opponent-Aware LLM Persuaders with Theory of Mind**](https://arxiv.org/pdf/2505.22961), which is trained with RL but without theory of mind information.

Please refer to our [Github Repo](https://github.com/Hanpx20/ToMAP) for usage details.