Annotated bibliography by ANNA VITORES GONZALEZ (2403303)

avitoresgonzalez — 2024-11-07 14:09:31 UTC

Resource 3

avitoresgonzalez — 2024-11-07 18:01:35 UTC

Lo, L.S. (2023) 'The CLEAR path: A framework for enhancing information literacy through prompt engineering', The Journal of Academic Librarianship, 49(4), pp. 102720. Available at: https://doi.org/10.1016/j.acalib.2023.102720

avitoresgonzalez — 2024-11-07 18:03:02 UTC

This paper from 2023 proposes the CLEAR (Concise, Logical, Explicit, Adaptative and Reflective) framework as a guide for creating effective prompts for generative AI (GenAI). The key elements of the framework are using only essential words, ordering ideas purposively, providing precise instruction about scope, content and format, reviewing and refining results, and thinking critically about the results. Although it doesn’t really teach how to do prompt engineering, it provides example prompts that use the framework, and indicates other technical aspects of prompts (tokens, temperature and top-p) which help the reader understand some of the things that influence the outputs obtained by GenAI. The author suggests librarians and educators integrate the framework into their information literacy practices, and refine and improve it by sharing their experiences with the broader community.

The paper has 95 citations in Scopus; Scimago ranks this journal 34^th out of 271 library and information science journals. The author has a 15-year track record of combining academic librarianship with active research.

avitoresgonzalez — 2024-12-01 13:29:07 UTC

This annotated bibliography aims to provide research insights and resources for information professionals working in academic environments. Advances in GenAI are a hot topic in the higher education sector, and the services provided by academic libraries are likely to be transformed due to the proliferation and popularisation of GenAI tools. Especially when thinking about information-seeking and retrieval, it is difficult to understate how GenAI tools are changing expectations about what searching for and accessing information means.

This selection includes research and resources aimed at understanding and developing the prompting skills required to use GenAI tools effectively when searching for information, alongside reflections, insights and examples of how to integrate critical appraisal of these tools and the skills needed to use them with integrity into information literacy programmes.

The selected resources were collected during October and November 2024, and they were highly relevant at that date, but it is important to highlight that GenAI, GenAI tools, and the study of their ethical and societal implications is a rapidly evolving field. It is advisable to monitor updates, developments and associated research on a regular basis.

avitoresgonzalez — 2024-12-01 13:33:04 UTC

This resource from the International Federation of Library Associations and Institutions (IFLA) offers a guide on GenAI for library and other information professionals. It includes a general introduction to GenAI and how it works, and sections on ethical and information issues, GenAI’s uses for information professionals, and the role of information professionals in guiding other ‘end users’ of GenAI tools. Each section contains high-level information and guidance, with links to a wide range of further resources for more in-depth treatment.

IFLA is an independent, international NGO, established in 1927 to advance the interests of libraries and other information services and associations, librarians and their communities. It is a highly respected organisation with a long track record of providing support, guidance, training and development to the information sectors on new and challenging issues.

Resource 1

avitoresgonzalez — 2024-12-01 13:45:38 UTC

IFLA (2024) Generative AI for library and information professionals. Available at: https://www.ifla.org/g/ai/generative-ai/ (Accessed: 2 November 2024).

Resource 2

avitoresgonzalez — 2024-12-01 13:47:00 UTC

Cox, A. (2024) 'Algorithmic Literacy, AI Literacy and Responsible Generative AI Literacy', Journal of Web Librarianship, pp. 1–18. Available at: https://doi.org/10.1080/19322909.2024.2395341

avitoresgonzalez — 2024-12-01 13:47:28 UTC

Published in October 2024, this paper explores how GenAI literacy could be defined by analysing its relations to other related literacies. Specifically, the paper reviews different approaches to “Algorithmic literacy” and “AI literacy”, highlighting key literature in this area that helps frame the question of what GenAI literacy should involve. It suggests that Gen AI literacy is constituted by theorical, ethical, pragmatic and reflective knowledge, and provides a useful table to help think about the different ways in which those four types of knowledge are activated through the process of using GenAI. The text provides a useful framework for developing and/or evaluating guidance for using GenAI, and also helps contextualize “prompt engineering” as just one part of GenAI literacy.

The author is an expert in the response of information professionals to new challenges, based at University of Sheffield's Information School, which has been ranked number one in the world for Library and Information Management in the QS World University Rankings for the past four years.

Resource 4

avitoresgonzalez — 2024-12-01 13:58:02 UTC

Chowdhury, G. and Chowdhury, S. (2024) ‘AI- and LLM-driven search tools: A paradigm shift in information access for education and research’ Journal of Information Science, Online first. Available at: https://doi.org/10.1177/01655515241284046

avitoresgonzalez — 2024-12-01 14:01:52 UTC

This paper from October 2024 reviews the outcomes of various search tasks conducted through interactions with six GenAI tools (Claude, Perplexity, Copilot, Gemini, GPT-4o, and Scholar GPT). Outcomes obtained through specific prompts are analysed and evaluated for their suitability and relevance for a diverse range of academic needs. This provides valuable insights about the need to engage with the right tools, and about how to approach interaction with those tools. It is not intended to be a comparative study of the tools, but it provides some indication of their differing performance in response to specific tasks. The authors’ final reflections invite us to explore how to adapt to these kinds of tools, to reap their benefits while ensuring that outcomes meet the standards of libraries services (e.g. quality, accuracy, consistency, adaptation to specific user needs).

The journal is peer-reviewed and ranked 61^st out of 271 library and information science journals in Scimago. The lead author of the study has an extensive track record of research focused on information services and digital libraries.

Resource 5

avitoresgonzalez — 2024-12-01 14:04:37 UTC

Research Libraries UK (2024) Prompt engineering in libraries: some early observations. Available at: https://www.youtube.com/watch?v=JUEUkMz_ibo (Accessed: 2 November 2024).

avitoresgonzalez — 2024-12-01 14:05:00 UTC

This is one of this year’s “RLUK Digital Shift Forum” seminars , in which professionals discuss digital challenges in the sector. In this seminar, librarians from the University of Essex reflect on “prompt engineering”, sharing how they approach training in this area, and signposting to helpful resources and their applications. Taking into account the crucial role of academic integrity, the seminar offers “hands-on” examples of how to integrate prompt engineering within general information literacy training, as well as advice on how to develop skills in the area. It introduces terminology (e.g. “system message prompting”, “context-expansion”) that could help to develop prompt-construction strategies. It also debates the pros and cons of developing “prompt libraries” as a resource in higher education libraries.

Research Libraries UK is a consortium of the leading research libraries in the UK and Ireland. The speakers bring a range of perspectives from different aspects of librarianship including senior leadership.

Resource 6

avitoresgonzalez — 2024-12-01 14:06:12 UTC

Bsharat, S.M., Myrzakhan, A. and Shen, Z. (2023) ‘Principled instructions are all you need for questioning LLaMA-1/2, GPT-3.5/4’. arXiv, arXiv2312.16171v2. Available at: https://arxiv.org/abs/2312.16171 (Accessed: 2 November 2024).

avitoresgonzalez — 2024-12-01 14:08:57 UTC

Recommended in the RLUK seminar (resource 5 in this bibliography), this paper published in 2023 on arXiv (a free distribution service for STEM subject articles) provides guidance to help develop prompting skills. It offers technical context for understanding how prompting works for large language models, and also specific examples to help develop skills when interacting with these systems. Although prompt engineering should not be approached in isolation from general information literacy and specific GenAI literacy skills and learning, the principles and categories provided seem useful to understand and develop prompting skills.

The paper has not been peer reviewed, but has been recommended by experts. It has 82 citations in Google Scholar. According to CSRankings (Computer Science Rankings), Mohamed bin Zayed University of Artificial Intelligence, where all three authors are based, is ranking 19^th globally for natural language processing.

Resource 7

avitoresgonzalez — 2024-12-01 14:12:46 UTC

University of Sheffield (2024) Using AI and GenAI to find information. Available at: https://www.sheffield.ac.uk/study-skills/research/searching/using-ai#using-ai-and-genai-to-find-information (Accessed: 3 November 2024).

avitoresgonzalez — 2024-12-01 14:13:02 UTC

This webpage provides access to the University of Sheffield’s online tutorial for students about using AI and GenAI for finding information, framed as critical GenAI literacy. The tutorial is organized in eight sections, starting with a general introduction and definitions, followed by advice on topics such as discovering GenAI tools, using GenAI for search planning, and understanding these tools in the context of academic literature. It also includes an extensive section on thinking critically about GenAI, covering perspectives and biases, reliability and misinformation, privacy and online safety, ownership and copyright, environmental impact, digital equity, and inclusivity. Each section also includes access to other resources, many of which are relevant for research, teaching and information staff as well as students, such as an “Academic GenAI Prompt Bank”.

The University of Sheffield's Information School has been ranked number one in the world for Library and Information Management in the QS World University Rankings for the past four years.

Resource 8

avitoresgonzalez — 2024-12-01 14:14:05 UTC

Carroll, A.J. and Borycz, J. (2024) 'Integrating large language models and generative artificial intelligence tools into information literacy instruction', Journal of Academic Librarianship, 50(4), 102899. Available at: https://doi.org/10.1016/j.acalib.2024.102899

avitoresgonzalez — 2024-12-01 14:14:23 UTC

This paper from July 2024 outlines the authors’ strategy for integrating GenAI into information literacy training for STEM students at their institution. It provides practical examples of how librarians can shape students’ understanding of GenAI tools, teaching if, how and when to use them, and raising awareness of limitations and issues, e.g. around accuracy and incompleteness. The authors provide access to their teaching materials and suggest that GenAI could be used for teaching traditional information retrieval skills such as designing search strings, which would leave more time for librarians to focus on developing critical thinking skills around information literacy. While the study acknowledges its limitations (e.g. limited student sample and lack of follow up data), it is a useful practical example of introducing GenAI tools in the context of information literacy instruction.

The lead author is the Associate Director of the Science and Engineering Library (SEL) at Vanderbilt University. His work has won multiple awards from the American Library Association and the American Society for Engineering Education. Scimago ranks this journal 34^th out of 271 library and information science journals.

Resource 9

avitoresgonzalez — 2024-12-01 14:17:30 UTC

Chaudhuri, J. and Terrones, L. (2024) ‘Reshaping academic library information literacy programs in the advent of ChatGPT and other generative AI technologies’, Internet Reference Services Quarterly, pp. 1–25. Available at: https://doi.org/10.1080/10875301.2024.2400132

avitoresgonzalez — 2024-12-01 14:17:52 UTC

This paper from September 2024 reports on three information literacy initiatives developed by librarians at the University Library at California State University, Los Angeles, focused on instructing students on the use of ChatGPT. The first initiative addressed plagiarism; the second, library research; and the third combined key aspects of the two previous, asynchronous initiatives into a recurring live online workshop. The data gathered from these initiatives helps to understand the ways in which students are using GenAI in their information seeking and other academic practices. By discussing the development and evaluation of the initiatives, the paper provides an example road-map for information professionals to support their own development of AI literacy programmes. It highlights the need for information professionals to guide students in using GenAI tools in library research ethically and with integrity, and outlines some key elements of information professionals’ roles in this regard.

The authors are both practicing instructional librarians with significant experience of delivering teaching and learning activities for students around information literacy, and have developed an approach to AI literacy that aims to support rather than punish its use.

Resource 10

avitoresgonzalez — 2024-12-01 14:20:27 UTC

Alshami, A. et al. (2023) ‘Harnessing the power of ChatGPT for automating systematic review process: Methodology, case study, limitations, and future directions. Systems, 11(7), 351. Available at: https://doi.org/10.3390/systems11070351

avitoresgonzalez — 2024-12-01 14:21:26 UTC

One area that GenAI is regarded as potentially useful is helping with systematic reviews. This study investigates the use of ChatGPT to enhance the process of conducting systematic reviews, developing a model to integrate ChatGPT into the process, and comparing outcomes (in terms of efficiency and accuracy) with traditional methods. It explores the use of ChatGPT for a) developing Boolean search terms and collecting articles, b) screening abstracts and categorizing papers, c) filtering the full-text and extracting information, and d) analysing content. Their findings show positive impacts from using ChatGPT to generate Boolean search terms, classify and filter articles, but found limitations in extracting information. Their descriptions of specific limitations could be helpful to assess future developments in this area.

The article has had 46 citations in Scopus and has been cited by the UK Government Behavioural Insights Team (in ‘A Blueprint for Better International Collaboration on Evidence’, 2024).

avitoresgonzalez — 2024-12-01 14:56:49 UTC

avitoresgonzalez — 2024-12-01 14:57:43 UTC

avitoresgonzalez — 2024-12-01 15:01:01 UTC

avitoresgonzalez — 2024-12-01 15:02:39 UTC

"Library" by ellen forsyth is licensed under CC BY-SA 2.0.

avitoresgonzalez — 2024-12-01 15:07:11 UTC