Automatically Search and Download Papers

This script demonstrates the usage of AutoSearch from the auto_research.search.core module to:

• Perform a search for research papers based on specified keywords.

• Retrieve a specified number of articles.

• Sort the results by relevance or date.

• Apply filters such as a date cutoff and a minimum score threshold (auto_research.search.core.AutoSearch.score_threshold).

• Save the downloaded articles and metadata to a specified folder.

This script assumes that:

• The destination folder for storing search results exists or will be created.

------Searching for the 1th keyword 'Diffusion models'------

Searching papers:   0%|          | 0/3 [00:00<?, ?it/s]
Searching papers:  33%|███▎      | 1/3 [00:04<00:09,  4.55s/it]
Searching papers:  67%|██████▋   | 2/3 [00:07<00:03,  3.85s/it]
Searching papers: 100%|██████████| 3/3 [00:11<00:00,  3.90s/it]
Searching papers: 100%|██████████| 3/3 [00:11<00:00,  3.96s/it]


Paper 1:
Title: Diffusion models: A comprehensive survey of methods and applications
Abstract:

Being the most cutting-edge generative methods, diffusion methods have shown
great advances in wide generation tasks. Among them, graph generation attracts
significant research attention for its broad application in real life. In our
survey, we systematically and comprehensively review on diffusion-based graph
generative methods. We first make a review on three mainstream paradigms of
diffusion methods, which are denoising diffusion probabilistic models,
score-based genrative models, and stochastic differential equations. Then we
further categorize and introduce the latest applications of diffusion models on
graphs. In the end, we point out some limitations of current studies and future
directions of future explorations. The summary of existing methods metioned in
this survey is in
https://github.com/zhejiangzhuque/Diffusion-based-Graph-Generative-Methods.
Combined Score: 31.775648148732735
Citation count: 1486
Year of publication: 2023
Publication venue: ACM Computing Surveys
Authors: L Yang, Z Zhang, Y Song, S Hong, R Xu, Y Zhao


Link: https://arxiv.org/pdf/2209.00796
ArXiv Link: http://arxiv.org/pdf/2401.15617v2
Downloading Diffusion models A comprehensive survey of methods and applications.pdf... with upper time limit: 10 seconds
Downloaded: Diffusion models A comprehensive survey of methods and applications.pdf.


Paper 2:
Title: Diffusion models in vision: A survey
Abstract:

Diffusion models have become a new SOTA generative modeling method in various
fields, for which there are multiple survey works that provide an overall
survey. With the number of articles on diffusion models increasing
exponentially in the past few years, there is an increasing need for surveys of
diffusion models on specific fields. In this work, we are committed to
conducting a survey on the graph diffusion models. Even though our focus is to
cover the progress of diffusion models in graphs, we first briefly summarize
how other generative modeling methods are used for graphs. After that, we
introduce the mechanism of diffusion models in various forms, which facilitates
the discussion on the graph diffusion models. The applications of graph
diffusion models mainly fall into the category of AI-generated content (AIGC)
in science, for which we mainly focus on how graph diffusion models are
utilized for generating molecules and proteins but also cover other cases,
including materials design. Moreover, we discuss the issue of evaluating
diffusion models in the graph domain and the existing challenges.
Combined Score: 26.558112382722783
Citation count: 1242
Year of publication: 2023
Publication venue: IEEE Transactions on Pattern Analysis and Machine Intelligence
Authors: FA Croitoru, V Hondru, RT Ionescu


Link: https://arxiv.org/pdf/2209.04747
ArXiv Link: http://arxiv.org/pdf/2304.01565v1
Downloading Diffusion models in vision A survey.pdf... with upper time limit: 10 seconds
Downloaded: Diffusion models in vision A survey.pdf.


Paper 3:
Title: Variational diffusion models
Abstract:

The problem of spin diffusion is studied numerically in one-dimensional
classical Heisenberg model using a deterministic odd even spin precession
dynamics. We demonstrate that spin diffusion in this model, like energy
diffusion, is normal and one obtains a long time diffusive tail in the decay of
autocorrelation function (ACF). Some variations of the model with different
coupling schemes and with anisotropy are also studied and we find normal
diffusion in all of them. A systematic finite size analysis of the Heisenberg
model also suggests diffusive spreading of fluctuation, contrary to previous
claims of anomalous diffusion.
Combined Score: 3.699350861975652
Citation count: 1034
Year of publication: 2021
Publication venue: arXiv.org
Authors: D Kingma, T Salimans, B Poole


Link: https://proceedings.neurips.cc/paper/2021/file/b578f2a52a0229873fefc2a4b06377fa-Paper.pdf
ArXiv Link: http://arxiv.org/pdf/1212.2829v1
Downloading Variational diffusion models.pdf... with upper time limit: 10 seconds
Downloaded: Variational diffusion models.pdf.

The above displays all paper with a combined score no less than 0.5
Metadata saved to search_results/metadata.json

Folder saved to search_results.zip
------Searching for the 2th keyword 'Generative modeling for computer vision'------

Searching papers:   0%|          | 0/3 [00:00<?, ?it/s]
Searching papers:  33%|███▎      | 1/3 [00:03<00:06,  3.13s/it]
Searching papers:  67%|██████▋   | 2/3 [00:06<00:02,  2.98s/it]
Searching papers: 100%|██████████| 3/3 [00:09<00:00,  2.99s/it]
Searching papers: 100%|██████████| 3/3 [00:09<00:00,  3.00s/it]

The above displays all paper with a combined score no less than 0.5
Metadata saved to search_results/metadata.json

Folder saved to search_results.zip

from auto_research.search.core import AutoSearch


def main() -> None:
    """
    Main function to execute the AutoSearch process.

    This function initializes the search parameters, creates an instance of the `AutoSearch` class,
    and runs the search to retrieve and store the articles.

    Parameters
    ----------
    None

    Returns
    -------
    None
    """

    # Keywords can be a string or a list of strings. If keywords is a string, then perform a
    # search over the single keyword. If keywords is a list of strings, then perform a
    # search over each keyword in the list. Same articles from the search based on
    # different keywords will not be downloaded twice.
    keywords = ["Diffusion models", "Generative modeling for computer vision"]

    # Number of articles to retrieve at maximum.
    num_results = 3

    # Choose one between "relevance" and "date" for Google Scholar search engine
    sort_by = "relevance"

    # Date cutoff is relevant only when sort_by is set as "date".
    date_cutoff = "2024-12-01"

    # Minimum score that the article must have to be downloaded.
    score_threshold = 0.5

    # Path where the downloaded articles and metadata will be stored.
    destination_folder = "search_results"

    # Initialize the AutoSearch instance with the specified parameters.
    paper_search = AutoSearch(
        keywords=keywords,
        num_results=num_results,
        sort_by=sort_by,
        date_cutoff=date_cutoff,
        score_threshold=score_threshold,
        destination_folder=destination_folder,
    )

    # Run the search to retrieve and store the articles.
    paper_search.run()


if __name__ == "__main__":
    main()