GithubHelp home page GithubHelp logo

arthur-heng / nlgraph Goto Github PK

View Code? Open in Web Editor NEW
89.0 89.0 3.0 10.52 MB

Official repository of "Can Language Models Solve Graph Problems in Natural Language?". NeurIPS 2023 (Spotlight)

License: MIT License

Python 100.00%

nlgraph's Introduction

  • 👋 Hi, I’m @Arthur-Heng
  • 👀 I’m interested in playing the piano and participating in programming competitions
  • 🌱 I’m currently conducting research in NLP & ML
  • 💞️ I'm looking for Ph.D. positions starting in 25Fall!
  • 📫 How to reach me : e-mail: [email protected] || [email protected] || [email protected]

nlgraph's People

Contributors

arthur-heng avatar bunsenfeng avatar

Stargazers

 avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar  avatar

Watchers

 avatar

nlgraph's Issues

Shortest_path task problems

When running the shortest_path tasks, I observed two problems:

  1. The model is unable to create route directly connecting neighbor nodes. Namely, if the questioned nodes are directly connected, the model will not take this route as the predicted answer. It will always find another longer route.
  2. The predicted answers connected nodes with weight = 0. Namely, sometimes the predicted route follows a route that doesn't exist.
    Is there an issue in model training, or is it because of some errors in my operation?
    If it's because of model training, is it possible to clarify in the prompt that "nodes with edge weighing 0 should not be connected"?

process stucked

Hello!
image
Do you know if this is normal? It has taken 5 minutes and still not moved on the bar. Thanks and I expect your response.

Big Graph can be handeled?

Could the Large Language Model solve bigger graphs, such as Cora or other standard graph settings, which have thousands or more nodes?

file details

Thank you for sharing these data.

I have a few questions regarding the details:

  1. Could you kindly clarify on what does the first line number represent in the 'graph*.txt' file?
5 1 2
2 3
2 3
3 2
4 3
4 1
  1. As per my understanding, 'main.json' appears to be similar to 'train.json' and 'test.json' in the context of supervised fine-tuning setting?

Maybe add a license

Dear Authors,
Could you please provide a license for this open-source dataset, is it the same as the published paper? Or another, like an MIT license?

Question about the answers of datasets

Dear authors,

Thank you for your amazing contributions to this filed. I'm wondering whether your genration results with cot or bag prompting skills will be released?

Best,
ZY

Recommend Projects

  • React photo React

    A declarative, efficient, and flexible JavaScript library for building user interfaces.

  • Vue.js photo Vue.js

    🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.

  • Typescript photo Typescript

    TypeScript is a superset of JavaScript that compiles to clean JavaScript output.

  • TensorFlow photo TensorFlow

    An Open Source Machine Learning Framework for Everyone

  • Django photo Django

    The Web framework for perfectionists with deadlines.

  • D3 photo D3

    Bring data to life with SVG, Canvas and HTML. 📊📈🎉

Recommend Topics

  • javascript

    JavaScript (JS) is a lightweight interpreted programming language with first-class functions.

  • web

    Some thing interesting about web. New door for the world.

  • server

    A server is a program made to process requests and deliver data to clients.

  • Machine learning

    Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.

  • Game

    Some thing interesting about game, make everyone happy.

Recommend Org

  • Facebook photo Facebook

    We are working to build community through open source technology. NB: members must have two-factor auth.

  • Microsoft photo Microsoft

    Open source projects and samples from Microsoft.

  • Google photo Google

    Google ❤️ Open Source for everyone.

  • D3 photo D3

    Data-Driven Documents codes.