Model Card for IDMGSP-Galactica-TRAIN-CG
A fine-tuned Galactica model to detect machine-generated scientific papers based on their abstract, introduction, and conclusion.
This model is trained on the
train-cg dataset found in
https://huggingface.co/datasets/tum-nlp/IDMGSP.
this model card is WIP, please check the repository, the dataset card and the paper for more details.
Model Details
Model Description
- Developed by: Technical University of Munich (TUM)
- Model type: [More Information Needed]
- Language(s) (NLP): English
- License: [More Information Needed]
- Finetuned from model [optional]: Galactica
Model Sources
- Repository: https://github.com/qwenzo/-IDMGSP
- Paper: [More Information Needed]
Uses
Direct Use
1from transformers import AutoTokenizer, OPTForSequenceClassification, pipeline
2
3model = OPTForSequenceClassification.from_pretrained("tum-nlp/IDMGSP-Galactica-TRAIN-CG")
4tokenizer = AutoTokenizer.from_pretrained("tum-nlp/IDMGSP-Galactica-TRAIN-CG")
5reader = pipeline("text-classification", model=model, tokenizer = tokenizer)
6reader(
7'''
8Abstract:
9....
10
11Introduction:
12....
13
14Conclusion:
15...'''
16)
Downstream Use [optional]
[More Information Needed]
Out-of-Scope Use
[More Information Needed]
Bias, Risks, and Limitations
[More Information Needed]
Training Details
Training Data
The training dataset comprises scientific papers generated by the Galactica, GPT-2, and SCIgen models, as well as papers extracted from the arXiv database.
The provided table displays the sample counts from each source utilized in constructing the training dataset.
The dataset could be found in
https://huggingface.co/datasets/tum-nlp/IDMGSP.
| Dataset | arXiv (real) | ChatGPT (fake) | GPT-2 (fake) | SCIgen (fake) | Galactica (fake) | GPT-3 (fake) |
|---|
| TRAIN without ChatGPT (TRAIN-CG) | 8k | - | 2k | 2k | 2k | - |
Training Procedure
Preprocessing [optional]
[More Information Needed]
Training Hyperparameters
[More Information Needed]
Speeds, Sizes, Times [optional]
[More Information Needed]
Evaluation
Testing Data, Factors & Metrics
Testing Data
[More Information Needed]
Factors
[More Information Needed]
Metrics
[More Information Needed]
Results
[More Information Needed]
Summary
Model Examination [optional]
[More Information Needed]
Environmental Impact
- Hardware Type: [More Information Needed]
- Hours used: [More Information Needed]
- Cloud Provider: [More Information Needed]
- Compute Region: [More Information Needed]
- Carbon Emitted: [More Information Needed]
Technical Specifications [optional]
Model Architecture and Objective
[More Information Needed]
Compute Infrastructure
[More Information Needed]
Hardware
[More Information Needed]
Software
[More Information Needed]
Citation [optional]
BibTeX:
[More Information Needed]
APA:
[More Information Needed]
Glossary [optional]
[More Information Needed]
More Information [optional]
[More Information Needed]
Model Card Authors [optional]
[More Information Needed]
Model Card Contact
[More Information Needed]
