Model Card for SWAP_LLM_v2

SWAP_LLM_v2 is a suite of supervised fine-tuned models developed for multi-step reasoning with large language models (LLMs). The framework encompasses two primary components: generator and discriminator.

Model Details

Generator

• Base Model: meta-llama/Meta-Llama-3-8B-Instruct

• LoRA Configuration:

  • lora_alpha: 32
  • r: 16
  • target_modules: ["up_proj", "down_proj", "gate_proj", "q_proj","k_proj", "v_proj", "o_proj"]
  • bias: "none"

Discriminator

• Base Model: meta-llama/Meta-Llama-3-8B-Instruct

• LoRA Configuration:

  • lora_alpha: 32
  • r: 16
  • target_modules: ["up_proj", "down_proj", "gate_proj", "q_proj","k_proj", "v_proj", "o_proj"]
  • bias: "none"

For additional information and implementation details, please refer to the SWAP GitHub repository.

Citation

@inproceedings{xiong2025deliberate,
  title={Deliberate reasoning in language models as structure-aware planning with an accurate world model},
  author={Xiong, Siheng and Payani, Ali and Yang, Yuan and Fekri, Faramarz},
  booktitle={Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)},
  pages={31900--31931},
  year={2025}
}