EagleW/PaperRobot
Code for PaperRobot: Incremental Draft Generation of Scientific Ideas
Combines relation extraction and sequence generation in a two-stage pipeline: link prediction over knowledge graphs extracted from PubMed identifies relevant entities and facts, while neural generation models produce paper sections (abstracts, conclusions, titles) conditioned on these predicted relations. Built on PyTorch with NetworkX for graph operations, the system trains on paired PubMed data covering both entity linking and text generation tasks.
483 stars. No commits in the last 6 months.
Stars
483
Forks
131
Language
Python
License
MIT
Category
Last pushed
Mar 09, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/nlp/EagleW/PaperRobot"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Related tools
batzner/tensorlm
Wrapper library for text generation / language models at character and word level with RNNs in TensorFlow
microsoft/CodeMixed-Text-Generator
This tool helps automatic generation of grammatically valid synthetic Code-mixed data by...
tallpauley/wordsiv
A Python library for generating text with a limited character set, with type proofing in mind
yingpengma/Awesome-Story-Generation
This repository collects an extensive list of awesome papers about Story Generation /...
Cyrilvallez/TextWiz
An even simpler way to generate text with LLMs.