MétaCan
Menu
Back to cohort

On the data requirements of probing

2022· article· en· W4226145103 on OpenAlexaff
Zining Zhu, Jixuan Wang, Bai Li, Frank Rudzicz

Bibliographic record

VenueFindings of the Association for Computational Linguistics: ACL 2022 · 2022
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsVector InstituteUniversity of Toronto
Fundersnot available
KeywordsComputer scienceReliability (semiconductor)Context (archaeology)Construct (python library)Data miningMachine learningArtificial intelligencePower (physics)

Abstract

fetched live from OpenAlex

As large and powerful neural language models are developed, researchers have been increasingly interested in developing diagnostic tools to probe them.There are many papers with conclusions of the form "observation X is found in model Y ", using their own datasets with varying sizes.Larger probing datasets bring more reliability, but are also expensive to collect.There is yet to be a quantitative method for estimating reasonable probing dataset sizes.We tackle this omission in the context of comparing two probing configurations: after we have collected a small dataset from a pilot study, how many additional data samples are sufficient to distinguish two different configurations?We present a novel method to estimate the required number of data samples in such experiments and, across several case studies, we verify that our estimations have sufficient statistical power.Our framework helps to systematically construct probing datasets to diagnose neural NLP models.

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)0.151
metaresearch head score (Gemma)0.685
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.151
Threshold uncertainty score0.801

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.1510.685
Meta-epidemiology (narrow)0.0020.004
Meta-epidemiology (broad)0.0060.003
Bibliometrics0.0030.008
Science and technology studies0.0040.010
Scholarly communication0.0110.034
Open science0.0100.014
Research integrity0.0110.014
Insufficient payload (model declined to judge)0.0120.005

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 head0.054
GPT teacher head0.291
Teacher spread0.237 · 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 designSimulation or modeling
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

Citations4
Published2022
Admission routes1
Has abstractyes

Explore more

Same venueFindings of the Association for Computational Linguistics: ACL 2022Same topicTopic ModelingFrench-language works237,207