A curated list of Graph/Transformer-based fraud and anomaly papers and resources with an interactive dashboard and a local RAG-based LLM chatbot; repo created 2019-11-21, last push 2026-06-29, 1880 stars, 298 forks, license not listed.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
A curated list of Graph/Transformer-based papers and resources for fraud, anomaly, and outlier detection. The project documents an interactive dashboard to view/filter/search the listed papers and mentions a local RAG-based LLM chatbot with access to 250 publicly available papers.
How it works
The README lists structural components: an interactive paper dashboard and a RAG-based LLM chatbot for paper access. It groups content into sections such as LLM and Transformer Papers and Deep Learning Graph Papers by year, plus Toolbox, Dataset, Survey Paper, and Other Resource categories. Specific implementation behavior or architecture details are not provided in the excerpt.
Getting started
Commands are shown in the README as part of the tooling references:
- The README references an interactive dashboard at https://safe-graph.github.io/paper_dashboard/.
- It also mentions a local RAG-based LLM chatbot project with repository https://github.com/YingtongDou/paper_chatbot and notes to refer to that project README on deployment.
Recent releases
There are no RELEASES (latest 0) sections listed, and the README excerpt does not include versioned release notes.
Traction
Stars: 1880 Forks: 298 (open_issues: 0) overall metrics are provided in the FACTS block. The README does not include additional traction data beyond these counts.
Behind the repo
The repo description identifies a focus on academic-publications, anomaly-detection, data-mining, graph methods, and transformer-based papers. It links to an interactive dashboard and mentions a related chatbot project and a dedicated paper dashboard website.
Caveats
License: none listed. Created: 2019-11-21. Last push: 2026-06-29. Language: not specified. No explicit release notes in the truncated README.