SCINDR - The SCience INtroDuction Robot that will Connect Open Scientists
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
This project will develop a way to connect, in real time, globally disparate researchers who are doing similar science so that they can work better and faster towards the development of new medicines. The scientific literature already fulfills the role of notifying researchers about work that has been done, and social media has recently evolved to alert researchers to what is being done. While these new communication technologies simplify the collaborative process between widely distributed researchers, there still exists a major gap in efficient real time alerting and updating. We aim to automate an alert process so that, as a researcher records what they are doing in a natural way, they are immediately alerted to others around the world in real time who are working on related science. Our system is built on the conceptual model of the machine understanding of human-generated content, used by social media platforms to generate alerts to further relevant content. The system we propose to build will understand the molecular information being recorded in a scientist’s notebook. It will then search both its own records and others in the public domain in order to introduce scientists where there may be mutual advantage - when two laboratories are working on similar molecules, assays or approaches, for example. To achieve this, we will build on a recently developed open source electronic lab notebook (ELN) to create the required component - the automated alerting service we call the SCience INtroDuction Robot, or SCINDR. We foresee wide application of SCINDR in chemical and biological research because it will accelerate research by connecting people. In so doing, SCINDR will provide the incentive for others to take their research into the public domain (Fig. 1).
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 | 0.004 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.000 | 0.001 |
| Open science | 0.001 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 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