The Rank-Order Method for Appellate Subset Selection
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
Appellate courts in many countries will often use a subset of the entire appellate body to decide cases. The United States courts of appeals, the European Court of Justice, and the highest courts in Canada, Israel, South Africa, New Zealand, and the United Kingdom all use subsets. In general, there have been two methods that appellate courts have used to choose their subsets: direct selection and random assignment. In direct selection, the chief judge or a designated court administrator simply selects the members and size of the panel for that particular case. In random assignment, the size of the panel is preset and the composition of the panel is randomly assigned from the full set of judges. Both of these subset selection methods likewise involve a trade-off. Direct selection allows for panels that reflect the views of the entire set of judges, but also permits the “gaming” of the outcome in particular cases. Random assignment prevents such purposeful gaming, but allows for non-representative outlier panels to form as a matter of simple probability. This Essay introduces a new method for selecting subsets that combines the best elements of both the direct selection method and random assignment, while avoiding their pitfalls. This new method — which I call the rank-order method — creates subsets that are judicially efficient and representative of the appellate body as a whole. Importantly, the rank-order method also mitigates against possible “judicial gaming.”
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 | 0.002 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.000 |
| Science and technology studies | 0.003 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.001 |
| 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