ARLIS/NA Multimedia & Technology Reviews is seeking reviewers for the
following AI tools and resources relevant to art librarianship and art
history.

To volunteer, choose a resource from the list below and complete our Reviewer
Interest form <https://forms.gle/8sVZsi9JyF4mfBDW9> by *Friday, August 1,
2025*.

If you are interested in reviewing a tool that is not listed here, you are
welcome to *nominate a resource* using the same form.

Initial draft submissions are due *Tuesday, September 2nd, 2025*.

Contributing to ARLIS/NA Multimedia & Technology Reviews
<https://multimediatechnologyreviews.arlisna.hcommons.org/> is a great
opportunity to get involved with the Society, learn about interesting new
resources, and help shape the conversation around new tools in our field.
Please read the see the reviewer guidelines for this issue
<https://multimediatechnologyreviews.arlisna.hcommons.org/reviewer-guidelines-october-2025-issue/>,
and direct comments and questions about the reviews to [log in to unmask]
.

Submitted by ARLIS/NA Multimedia & Technology Reviews Editors:
Karina Wratschko
Matthew Garklavs
Abigail Walker

*__________________________________________*

*Resources for Review*The brief descriptions below are excerpted from each
resource’s official website and do not necessarily reflect the views
of the *M&T
Reviews* Editors.

   1. *Annif *https://annif.org/
   Automated subject indexing toolkit, a tool from the National Library of
   Finland which is designed for automated subject cataloging. Annif provides
   access to multiple ML backends facilitating trials of different ML models
   and approaches, including term frequency - inverse document frequency
   (TF-IDF) and multi-modal language model (MLLM), and benchmarking a wide
   range of approaches to subject and genre cataloging
   2. *EBSCO's AI-Enhanced Search*
   Includes AI Insights
   <https://www.ebsco.com/artificial-intelligence/products/ai-insights>
and Natural
   Language Search
   <https://www.ebsco.com/artificial-intelligence/products/natural-language-search>
(NLS). AI Insights
   generates a short list of key insights from full-text articles, while NLS
   improves the researcher’s experience when using their natural language,
   enhancing context handling and overall search accuracy.
   3. *ProQuest Research Assistant*
   https://support.proquest.com/s/article/ProQuest-Research-Assistant-FAQs?language=en_US
ProQuest Research Assistant integrates AI-powered capabilities into
   research workflows, offering suggestions and prompts at critical points. It
   helps users formulate next steps and boosts productivity by providing
   document insights instead of simple generative answers to research
   questions. With thousands of colleges and universities providing access to
   ProQuest Central, institutions can leverage AI to enhance academic outcomes
   while maintaining traditional scholarly standards.
   4. *JSTOR’s AI research tool *
   https://about.jstor.org/research-tool/#about
   JSTOR’s new interactive research tool.
   5. *WhisperAI*  https://github.com/openai/whisper
   Whisper is a general-purpose speech recognition model. It is trained on
   a large dataset of diverse audio and is also a multitasking model that can
   perform multilingual speech recognition, speech translation, and language
   identification.
   6. *Transkribus* https://www.transkribus.org/
   Transkribus enables you to automatically recognize text easily, edit
   seamlessly, collaborate effortlessly, and even train your custom AI for
   digitizing and interpreting historical documents of any form. *Comparative
   tool: Google Cloud OCR.*
   7. *Art Recognition* https://art-recognition.com/
   Art Recognition applies machine learning and computer vision techniques
   to determine the authenticity of artworks. We use two types
   of artificial neural networks to capture an artist’s main characteristics:
   a Convolutional Neural Network (CNN) and a Vision Transformer with Shifted
   Windows (SWIN). These cutting-edge architectures have been adapted for our
   art-classification task, to categorize artworks into “authentic” and
   “non-authentic.” While CNNs have conventionally excelled in image
   classification, transformers are a modern architecture that also drives
   applications such as ChatGPT. Art Recognition has led the way in adapting
   Vision Transformers for art authentication.
   8. *ArtPI* https://www.artpi.co/
   ArtPI is the first public API designed and optimized for art. It
   uses AI (artificial intelligence) and deep learning models trained over 1
   million artworks. ArtPI is aware of the concepts of style, genre, subject
   matter, composition, light, space, color and other principles and elements
   of art. ArtPI provides solutions for art market professionals and
   businesses, such as galleries, auction houses, museums, and online
   marketplaces. Individuals can search a database of public artworks at
   artpi.io while businesses can benefit from our easy-to-implement website
   plugin or full API to search their private collections and display a visual
   search options on their website along with "related content" predictions
   and recommendations to optimize visitor engagement and sales. Additionally,
   ArtPI can be used on social media networks, such as Instagram, to report on
   most shared artworks from an art fair or a gallery exhibit, just by drag
   and dropping a picture of the displayed artwork.
   9. *Scite *https://scite.ai/assistant
   Scite is an AI-powered research tool that helps researchers better
   discover and evaluate scientific literature through Smart Citations—a
   revolutionary system that shows whether articles support, contrast, or
   simply mention a given claim. Founded in 2018, and now part of Research
   Solutions, Scite has indexed over 1.3 billion citations and partnered with
   more than 30 major publishers to provide researchers with unparalleled
   access to scientific literature. With its Scite Assistant, Smart Citation
   Index, and advanced search capabilities, the platform addresses critical
   challenges such as information overload and research reproducibility.
   Trusted by two million active users worldwide, Scite is reshaping how
   researchers interact with scholarly content—building ethical,
   transparent AI tools that support rigorous, copyright-compliant research.
   10. *Living Museum App* https://www.livingmuseum.app/explore
   The Living Museum is an experimental user interface using content from
   the British Museum’s website[1] (unaffiliated), which aims to show how
   these technologies could be used to craft engaging experiences for museums
   visitors and people at home. First, we allow visitors to curate
   personalized exhibits by searching the collection using natural language.
   Second, we use large language models (LLMs) to bring artifacts to life,
   allowing visitors to feel their presence and learn about history through
   dialogue.


-- 
_____________________________________________________________
*Matthew Garklavs* | Electronic Resources Librarian, Assistant Professor
Pronouns: He/Him/His

<https://pratt.libcal.com/appointments?u=74286>
Schedule an appointment with me on Zoom
<https://pratt.libcal.com/appointments?u=74286>

*PRATT INSTITUTE*
200 Willoughby Avenue | Pratt Library 1 | Brooklyn, NY 11205
phone: (718) 399-4423 | [log in to unmask]
<https://www.pratt.edu/pratt-shows/pratt-shows-2024?utm_source=Signature&utm_medium=Digital&utm_campaign=Pratt-Shows-2024&utm_content=signature>