Why this work is in the frame
A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.
Bibliographic record
Abstract
ResearchRabbit is a scholarly publication discovery tool supported by artificial intelligence (AI).It was developed in 2021 by a team of three in Seattle [1].This tool lets users discover publications related to one or more seed publications with the help of visualization maps and lists of earlier, later, and similar publications.ResearchRabbit is designed to support the workflow of unstructured searching while providing a left-to-right trail from the original publication(s) through any selected authors or publications.These trails, which can run as deep as rabbit holes, suggest the origin of the tool's name.To start using ResearchRabbit, users first need to create an account.Then they need to create a collection and add at least one publication.The more publications that are added, the better ResearchRabbit can understand users' interests and generate recommendations similar to the contents of the collection.Publications can be added either by uploading a RIS or BibTeX file or by using ResearchRabbit's search, powered by PubMed, if users are searching the medical sciences, or Semantic Scholar, for any other subject area.While ResearchRabbit uses PubMed's and Semantic Scholar's search engines, the company claims its unique database of "100s of millions of academic articles" is second in size only to Google Scholar [2].Once publications are in a collection, ResearchRabbit's algorithm will begin generating recommendations.These recommendations can be explored through two modes: 1) by Papers that are Similar work, Earlier work, or Later work or 2) by People that provide additional publications that These authors or Suggested authors have published (Figure 1).These recommendations are depicted using visualization maps.
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.
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.098 | 0.182 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.005 |
| Science and technology studies | 0.003 | 0.000 |
| Scholarly communication | 0.007 | 0.005 |
| Open science | 0.004 | 0.000 |
| Research integrity | 0.001 | 0.006 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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 it