Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,62 @@
|
|
| 1 |
-
---
|
| 2 |
-
license: apache-2.0
|
| 3 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
license: apache-2.0
|
| 3 |
+
language: en
|
| 4 |
+
tags:
|
| 5 |
+
- persuasion
|
| 6 |
+
- llm
|
| 7 |
+
- dataset-generation
|
| 8 |
+
- multi-agent
|
| 9 |
+
- dialogue
|
| 10 |
+
- nlp
|
| 11 |
+
---
|
| 12 |
+
|
| 13 |
+
# 🧠 LLM-based Persuasion Simulator
|
| 14 |
+
|
| 15 |
+
Official model and repository for the following papers:
|
| 16 |
+
|
| 17 |
+
> **Communication Makes Perfect: Persuasion Dataset Construction via Multi-LLM Communication**
|
| 18 |
+
> *NAACL 2025 (Main Conference)*
|
| 19 |
+
> [https://aclanthology.org/2025.naacl-long.203/](https://aclanthology.org/2025.naacl-long.203/)
|
| 20 |
+
|
| 21 |
+
> **Communication is All You Need: Persuasion Dataset Construction via Multi-LLM Communication**
|
| 22 |
+
> *arXiv preprint, February 2025*
|
| 23 |
+
> [arXiv:2502.08896](https://arxiv.org/abs/2502.08896)
|
| 24 |
+
|
| 25 |
+
Both versions describe the same multi-LLM communication framework for generating persuasive dialogue datasets,
|
| 26 |
+
with the NAACL 2025 paper being the peer-reviewed and finalized version.
|
| 27 |
+
|
| 28 |
+
---
|
| 29 |
+
|
| 30 |
+
## 📘 Overview
|
| 31 |
+
|
| 32 |
+
This repository provides the **LLM-based persuasion simulation framework** introduced in our paper.
|
| 33 |
+
The framework enables automated generation of high-quality persuasive dialogues through **multi-agent communication among large language models (LLMs)**,
|
| 34 |
+
covering diverse linguistic and strategic dimensions of persuasion.
|
| 35 |
+
|
| 36 |
+
**Key features:**
|
| 37 |
+
- Multi-LLM communication to enhance dialogue fluency, coherence, and diversity
|
| 38 |
+
- Fine-grained control over persuasion strategies
|
| 39 |
+
- Modular design for flexible extension and reproducibility
|
| 40 |
+
- Applied to construct the *PersuaSim* and *PersuaSim-Culture* datasets
|
| 41 |
+
|
| 42 |
+
---
|
| 43 |
+
|
| 44 |
+
## 📂 Code Repository
|
| 45 |
+
|
| 46 |
+
💻 **Full implementation and scripts are available at:**
|
| 47 |
+
👉 [https://github.com/HF-heaven/LLM-based_persuasion_simulator](https://github.com/HF-heaven/LLM-based_persuasion_simulator)
|
| 48 |
+
|
| 49 |
+
---
|
| 50 |
+
|
| 51 |
+
## 🧩 Citation
|
| 52 |
+
|
| 53 |
+
If you find this work useful, please cite:
|
| 54 |
+
|
| 55 |
+
```bibtex
|
| 56 |
+
@inproceedings{ma2025communication,
|
| 57 |
+
title={Communication Makes Perfect: Persuasion Dataset Construction via Multi-LLM Communication},
|
| 58 |
+
author={Ma, Weicheng and Zhang, Hefan and Yang, Ivory and Ji, Shiyu and Chen, Joice and Hashemi, Farnoosh and Mohole, Shubham and Gearey, Ethan and Macy, Michael and Hassanpour, Saeed and others},
|
| 59 |
+
booktitle={Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 1: Long Papers)},
|
| 60 |
+
pages={4017--4045},
|
| 61 |
+
year={2025}
|
| 62 |
+
}
|