README.md

SurveyGen

Codes

1. Data Processing:

• Build Citation Network for Survery papers;
• Extract taxonomy trees;
• Reference lables (sub-community) for subsection / sub-topic; (Yuntong Hu)

2. Deep Clustering:

• GNNs;
• Hierarchical Clustering

3. LLM Generation:

• Prompt design;
• Generator;

Paper Writing

• Introduction;
• Related Work;
• Problem Formulation;
• Experiment;
• Conclution;
• Appendix.

---

Environment setup

conda create --name autosurvey python=3.9 -y
conda activate autosurvey

conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia
pip install torch_geometric
pip install backoff
pip install scholarly
pip install fuzzywuzzy
pip install nltk
pip install rank-bm25

---

Process dataset

python -m src.dataset.build_tree
python -m src.dataset.update_tree_abstract