Human-AI collaboration to identify literature for evidence synthesis
Bibliographic record
Abstract
Systematic approaches to evidence synthesis can improve the rigor, transparency, and replicability of a literature review. However, these systematic approaches are resource intensive. We evaluate the ability of ChatGPT to undertake two stages of evidence syntheses (searching peer-reviewed literature and screening for relevance) and develop a collaborative framework to leverage both human and AI intelligence. Using a scoping review of community-based fisheries management as a case study, we find that with substantial prompting, the AI can provide critical insight into the construction and content of a search string. Thereafter, we evaluate five strategies for synthesizing AI output to screen articles based on predefined inclusion criteria. We find that low omission rates (<1%) of relevant literature by the AI are achievable, which is comparable to human screeners. These findings suggest that generalized AI tools can assist reviewers to accelerate the implementation and improve the reliability of literature reviews, thus supporting evidence-informed decision-making.
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 | 0.609 | 0.751 |
| Meta-epidemiology (narrow) | 0.004 | 0.004 |
| Meta-epidemiology (broad) | 0.008 | 0.006 |
| Bibliometrics | 0.054 | 0.028 |
| Science and technology studies | 0.008 | 0.006 |
| Scholarly communication | 0.019 | 0.016 |
| Open science | 0.008 | 0.033 |
| Research integrity | 0.007 | 0.006 |
| Insufficient payload (model declined to judge) | 0.043 | 0.010 |
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; the direct Gemma label and the distilled Codex classifier agree on what is shown here.
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".