What can Artificial Intelligence tools currently do? And how might that affect work in academic libraries? Using the example of Elicit.org.
Bibliographic record
Abstract
This presentation provides a brief introduction to some of the basics of so-called Artificial Intelligence (AI) and its current fields of use. Based on a number of example applications, a non-exhaustive overview of the current market is provided to show in which areas AI tools already exist. The presentation then turns its spotlight onto the bibliographic discovery tool Elicit.org. This is an AI tool for identifying scholarly literature based on research questions that are submitted to the application and converted into search strings using Natural Language Processing (NLP). A list of matches relevant to the research question is then generated from the Semantic Scholar corpus of academic literature using the GPT-3 large language model. The retrieved papers are summarized and critically evaluated by the tool. At the end of the presentation, some shortcomings as well as potential library use cases of this application and of AI tools in general are discussed in order to encourage a critical but measured reflection on these instruments. The presentation is aimed to be an entry-level contribution to building AI literacy.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.011 | 0.029 |
| Meta-epidemiology (narrow) | 0.001 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.007 | 0.016 |
| Science and technology studies | 0.007 | 0.005 |
| Scholarly communication | 0.035 | 0.026 |
| Open science | 0.002 | 0.009 |
| Research integrity | 0.005 | 0.002 |
| Insufficient payload (model declined to judge) | 0.029 | 0.017 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".