Generating reading comprehension items using automated processes
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
Over the last five years, tremendous strides have been made in advancing the AIG methodology required to produce items in diverse content areas. However, the one content area where enormous problems remain unsolved is language arts, generally, and reading comprehension, more specifically. While reading comprehension test items can be created using many different item formats, fill-in-the-blank remains one of the most common when the goal is to measure inferential knowledge. Currently, the item development process used to create fill-in-the-blank reading comprehension items is time-consuming and expensive. Hence, the purpose of the study is to introduce a new systematic method for generating fill-in-the-blank reading comprehension items using an item modeling approach. We describe the use of different unsupervised learning methods that can be paired with natural language processing techniques to identify the salient item models within existing texts. To demonstrate the capacity of our method, 1,013 test items were generated from 100 input texts taken from fill-in-the-blank reading comprehension items used on a high-stakes college entrance exam in South Korea. Our validation results indicated that the generated items produced higher semantic similarities between the item options while depicting little to no syntactic differences with the traditionally written test items.
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 | 0.001 | 0.001 |
| 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 | 0.001 |
| Open science | 0.001 | 0.001 |
| 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