An algorithm for learning phonological classes from distributional similarity
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
An important question in phonology is to what degree the learner uses distributional information rather than substantive properties of speech sounds when learning phonological structure. This paper presents an algorithm that learns phonological classes from only distributional information: the contexts in which sounds occur. The input is a segmental corpus, and the output is a set of phonological classes. The algorithm is first tested on an artificial language, with both overlapping and nested classes reflected in the distribution, and retrieves the expected classes, performing well as distributional noise is added. It is then tested on four natural languages. It distinguishes between consonants and vowels in all cases, and finds more detailed, language-specific structure. These results improve on past approaches, and are encouraging, given the paucity of the input. More refined models may provide additional insight into which phonological classes are apparent from the distributions of sounds in natural languages.
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 | 0.001 | 0.000 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.005 | 0.001 |
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