MétaCan
Menu
Back to cohort

An information theoretic view on selecting linguistic probes

2020· article· en· W3099658661 on OpenAlexafffund
Zining Zhu, Frank Rudzicz

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicNatural Language Processing Techniques
Canadian institutionsVector InstituteUniversity of Toronto
FundersCanadian Institute for Advanced Research
KeywordsClassifier (UML)Artificial intelligenceModuloComputer scienceInformation gainNatural language processingSelection (genetic algorithm)Machine learningMathematicsDiscrete mathematics

Abstract

fetched live from OpenAlex

There is increasing interest in assessing the linguistic knowledge encoded in neural representations.A popular approach is to attach a diagnostic classifier -or "probe" -to perform supervised classification from internal representations.However, how to select a good probe is in debate.Hewitt and Liang (2019) showed that a high performance on diagnostic classification itself is insufficient, because it can be attributed to either "the representation being rich in knowledge", or "the probe learning the task", which Pimentel et al. (2020) challenged.We show this dichotomy is valid informationtheoretically.In addition, we find that the methods to construct and select good probes proposed by the two papers, control task (Hewitt and Liang, 2019) and control function (Pimentel et al., 2020), are equivalent -the errors of their approaches are identical (modulo irrelevant terms).Empirically, these two selection criteria lead to results that highly agree with each other.

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)0.019
metaresearch head score (Gemma)0.092
Version: metacan-v3-hybrid-931329e0061cValidation status: machine_predicted_unvalidated
Candidate categoriesnone
Consensus categoriesnone
DomainCandidate signal: none · Consensus signal: none
Study designCandidate signal: Theoretical or conceptual · Consensus signal: Theoretical or conceptual
GenreCandidate signal: Empirical · Consensus signal: none
Teacher disagreement score0.019
Threshold uncertainty score0.098

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0190.092
Meta-epidemiology (narrow)0.0020.001
Meta-epidemiology (broad)0.0020.002
Bibliometrics0.0050.003
Science and technology studies0.0020.015
Scholarly communication0.0070.020
Open science0.0050.006
Research integrity0.0060.006
Insufficient payload (model declined to judge)0.0080.001

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.

Opus teacher head0.011
GPT teacher head0.269
Teacher spread0.258 · 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 designTheoretical or conceptual
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

Citations14
Published2020
Admission routes2
Has abstractyes

Explore more

Same topicNatural Language Processing TechniquesFrench-language works237,207