MétaCan
Menu
Back to cohort
Record W2949241294 · doi:10.1145/3341692

Approximate normalization for gradual dependent types

2019· article· en· W2949241294 on OpenAlex

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.

affAt least one author lists a Canadian institution in the pinned OpenAlex snapshot.
fundA Canadian funder is recorded on the work.

Bibliographic record

VenueProceedings of the ACM on Programming Languages · 2019
Typearticle
Languageen
FieldComputer Science
TopicLogic, programming, and type systems
Canadian institutionsUniversity of British Columbia
FundersFP7 Ideas: European Research CouncilComisión Nacional de Investigación Científica y TecnológicaNatural Sciences and Engineering Research Council of Canada
KeywordsNormalization (sociology)SoundnessDecidabilityType inferenceStatic analysisType (biology)Program analysisType theoryData type

Abstract

fetched live from OpenAlex

Dependent types help programmers write highly reliable code. However, this reliability comes at a cost: it can be challenging to write new prototypes in (or migrate old code to) dependently-typed programming languages. Gradual typing makes static type disciplines more flexible, so an appropriate notion of gradual dependent types could fruitfully lower this cost. However, dependent types raise unique challenges for gradual typing. Dependent typechecking involves the execution of program code, but gradually-typed code can signal runtime type errors or diverge. These runtime errors threaten the soundness guarantees that make dependent types so attractive, while divergence spoils the type-driven programming experience. This paper presents GDTL, a gradual dependently-typed language that emphasizes pragmatic dependently-typed programming. GDTL fully embeds both an untyped and dependently-typed language, and allows for smooth transitions between the two. In addition to gradual types we introduce gradual terms , which allow the user to be imprecise in type indices and to omit proof terms; runtime checks ensure type safety. To account for nontermination and failure, we distinguish between compile-time normalization and run-time execution: compile-time normalization is approximate but total, while runtime execution is exact , but may fail or diverge. We prove that GDTL has decidable typechecking and satisfies all the expected properties of gradual languages. In particular, GDTL satisfies the static and dynamic gradual guarantees: reducing type precision preserves typedness, and altering type precision does not change program behavior outside of dynamic type failures. To prove these properties, we were led to establish a novel normalization gradual guarantee that captures the monotonicity of approximate normalization with respect to imprecision.

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 imitation

Not 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.

metaresearch head score (Codex)0.001
metaresearch head score (Gemma)0.000
Version: codex-gemma-dda1882f352aValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.649
Threshold uncertainty score0.435

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0010.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.000
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0020.001
Research integrity0.0000.000
Insufficient payload (model declined to judge)0.0000.000

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.

Opus teacher head0.017
GPT teacher head0.258
Teacher spread0.241 · how far apart the two teachers sit on this one work
Validation statusscore_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from it