Semantic Scholar

First check your course guide to see whether (and in what way) you may use GenAI in your course. In case there is no information in your course guide, using AI to generate the content of your assignments is considered fraud.

Semantic Scholar (semanticscholar.org) is an example of an AI tool that can be used for literature research.

Tools like Semantic Scholar are helpful for gaining a quick, overview understanding of a topic, but do not rely solely on them to provide you with all of the available information. Always use them along a regular systematic database search. Make sure to double check all provided information as well.

Semantic Scholar is free and open access, and has a database of over 200 million papers in all fields of science. Its search function uses AI to provide you with relevant papers, and has filters for among others publication date range, journal, and author, and the option to sort by e.g. date or citation count. Besides this it has multiple functionalities that help you gain a quick overview understanding of a topic. Note that BETA functions are only available for a limited number of papers.

  • On many papers (currently still mainly in the biomedical and computer science fields), it provides TLDRs (Too Long; Didn't Read) - these are super short summaries, generated using AI, that contain information on the paper's main objectives and results.
  • Highly Influential Citations: Semantic Scholar's models are able to determine which papers had a large influence on the paper you're interested in, which can help determine which papers to read fully.
  • Paper Recommendations: Once you've saved some interesting papers in a folder, the app provides you with personalised recommendations for similar papers. You can also turn on alerts to get notified when there's new relevant papers or citations.
  • Semantic Reader (BETA) is an 'AI-augmented' PDF reader. An example of its functionalities is that it allows you to see citation information without losing your place in the text. Besides this, it has a skimming mode where a paper's key points are highlighted, ensuring you can quickly find the main points of the paper. Its Generative Term Understanding function uses GenAI to generate explanations of important terms in the paper.
  • Ask This Paper (BETA) allows you to ask questions about a specific paper, that are subsequently answered using GenAI. Do ensure you ask questions that your paper is able to answer (e.g., 'What are the key results of this paper?').
  • Topic pages (BETA) are pages about important concepts mentioned in several papers. They include, for example, an AI-generated definition of the topic, related topics, and relevant papers.

See some examples of using Semantic Scholar below.

Example: A paper's page with TLDR and citation information

Figure 8

A screenshot of the Semantic Scholar page of a paper about wildfires, with arrows pointing to its TLDR and citation information.

Example: Skimming highlights in Semantic Reader

Figure 9

A screenshot of a Semantic Reader page, where key points are highlighted in different colours.

Example: Ask This Paper

Figure 10

A screenshot of the Ask This Paper functionality, where the question "What is the goal of this paper?" is asked.