Improving calibration time and accuracy for situation-specific models of color differentiation
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
Color vision deficiencies (CVDs) cause problems in situations where people need to differentiate the colors used in digital displays. Recoloring tools exist to reduce the problem, but these tools need a model of the user's color-differentiation ability in order to work. Situation-specific models are a recent approach that accounts for all of the factors affecting a person's CVD (including genetic, acquired, and environmental causes) by using calibration data to form the model. This approach works well, but requires repeated calibration - and the best available calibration procedure takes more than 30 minutes. To address this limitation, we have developed a new situation-specific model of human color differentiation (called ICD-2) that needs far fewer calibration trials. The new model uses a color space that better matches human color vision compared to the RGB space of the old model, and can therefore extract more meaning from each calibration test. In an empirical comparison, we found that ICD-2 is 24 times faster than the old approach, and had small but significant gains in accuracy. The efficiency of ICD-2 makes it feasible for situation-specific models of individual color differentiation to be used in the real world.
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 | 0.000 | 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 | 0.000 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 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