The Libraries are testing a new natural-language search tool and we need your help to make the service better.
The OneSearch Research Assistant beta is a tool powered by Generative Artificial Intelligence (specifically, Large Language Models or LLM). It allows you explore library content by asking questions in natural language. The Research Assistant uses content found in OneSearch to identify five documents that can help answer your question. It then extracts the most relevant information from the description/abstracts of each source to write the answer. Below the answer, you’ll see the sources used to generate it along with in-line citations that let you clearly see which source was used to generate each part in the answer. Use these sources to delve deeper into the topic, to restate your question, or to search in OneSearch for more sources.
How to Connect to the Research Assistant
You can access the Research Assistant beta in several ways:
- Navigate directly to https://onesearch.uark.edu/discovery/researchAssistant?vid=01UARK_INST:01UARK . You will be prompted to login to your library account.
- In OneSearch, look for the “sparkles” icon on the upper right of the screen. You will be prompted to login to your library account.
- In OneSearch, after logging into your library account, perform a search. In the results list, look for the Research Assistant link in the sidebar
What kinds of research questions can I ask?
To make the most of the Research Assistant beta, it’s essential to ask clear and detailed questions about academic or scientific topics. Be as specific as possible and – pretend you’re on Jeopardy! – phrase your query in the form of a question.
Example queries can be found on the starting screen and include research questions such as “How can we improve diversity in clinical trials? ” and “What are the most effective individual actions to reduce carbon emissions?”
The Research Assistant beta is not a chat bot. It does not yet support follow-up questions. Each question stands by itself. For example, if you ask “what is the most important work of Simone de Beauvoir”, you cannot follow up by asking “and what is the content of that work” and expect the system to understand what you mean. At this time, you will have to include all relevant information in each question, e.g. “what is the content of Beauvoir’s The Second Sex”?
The Research Assistant beta cannot yet refine your results by facets such as “peer-reviewed articles” or “articles published in the last 5 years.”
What sources are used to answer my question?
The Research Assistant beta bases its answers for the most part on online articles, book chapters, and theses. The Research Assistant Large Language Model does not include content from print books, manuscript collections, or other locally-held materials. Some publishers have opted out of including their information in this beta phase.
How does it work?
Your question is converted into a query that the search engine understands with the help of a Large Language Model (currently GPT 3.5). The search engine then identifies the most relevant documents in the index. It ranks them according to how well they can answer the question and, again a with the help of the Large Language Model, creates an answer from the top 5 sources.
Due to the nature of Large Language Models, answers to the same question are not always the same. There may be more than one possible answer and different resources that are relevant. If you are not satisfied with your answers, use the “Try again” button. Or try one of the suggested related research questions.
Is my information shared with the vendor?
Your personal data is not stored.
Anonymized data, including our institution name and questions asked, are kept for statistical analysis.
Answer results are not stored from session to session. However, your questions and feedback are used to train the Research Assistant to better respond to research queries.
Read more about the vendor’s privacy policies.
I have feedback!
Great! Use the thumbs up / thumbs down icons in your results to give us your comments, questions, and suggestions for improvement. We want your honest opinion on the usefulness of this new AI tool.
You are also welcome to send feedback directly to libweb@uark.edu with the subject line “OneSearch Research Assistant.”