MétaCan
Menu
Back to cohort

An Analysis of Dataset Overlap on Winograd-Style Tasks

2020· preprint· en· W3099859597 on OpenAlexafffund
Ali Emami, Kaheer Suleman, Adam Trischler, Jackie Chi Kit Cheung

Bibliographic record

Venuenot available
Typepreprint
Languageen
FieldComputer Science
TopicNatural Language Processing Techniques
Canadian institutionsMcGill University
FundersNatural Sciences and Engineering Research Council of CanadaCanadian Institute for Advanced ResearchMicrosoft Research
KeywordsSchema (genetic algorithms)Computer sciencePronounArtificial intelligenceNatural language processingStyle (visual arts)ParsingLanguage modelMachine learningLinguistics

Abstract

fetched live from OpenAlex

The Winograd Schema Challenge (WSC) and variants inspired by it have become important benchmarks for common-sense reasoning (CSR).Model performance on the WSC has quickly progressed from chance-level to near-human using neural language models trained on massive corpora.In this paper, we analyze the effects of varying degrees of overlap between these training corpora and the test instances in WSC-style tasks.We find that a large number of test instances overlap considerably with the corpora on which state-of-the-art models are (pre)trained, and that a significant drop in classification accuracy occurs when we evaluate models on instances with minimal overlap.Based on these results, we develop the KNOWREF-60K dataset, which consists of over 60k pronoun disambiguation problems scraped from web data.KNOWREF-60K is the largest corpus to date for WSC-style common-sense reasoning and exhibits a significantly lower proportion of overlaps with current pretraining corpora.

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.

How this classification was reachedexpand

Full frame machine prediction

Teacher imitation

Not calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.

metaresearch head score (Codex)0.011
metaresearch head score (Gemma)0.050
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Observational · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.011
Threshold uncertainty score0.060

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0110.050
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0020.001
Bibliometrics0.0030.005
Science and technology studies0.0020.002
Scholarly communication0.0030.004
Open science0.0020.006
Research integrity0.0020.002
Insufficient payload (model declined to judge)0.0030.002

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 head0.027
GPT teacher head0.330
Teacher spread0.303 · 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

Classification

machine, unvalidated

Machine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.

The models applied no category: nothing in the taxonomy fit this work.
Study designObservational
Domainnot available
GenreEmpirical

How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".

Quick stats

Citations1
Published2020
Admission routes2
Has abstractyes

Explore more

Same topicNatural Language Processing TechniquesFrench-language works237,207