MétaCan
Menu
Back to cohort
Record W4280612849 · doi:10.18653/v1/2022.acl-short.45

Problems with Cosine as a Measure of Embedding Similarity for High Frequency Words

2022· article· en· W4280612849 on OpenAlexfundno aff
Kaitlyn Zhou, Kawin Ethayarajh, Dallas Card, Dan Jurafsky

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsnot available
FundersNatural Sciences and Engineering Research Council of CanadaNational Science Foundation
KeywordsCosine similaritySimilarity (geometry)PolysemyWord (group theory)Natural language processingTrigonometric functionsEmbeddingComputer scienceArgument (complex analysis)Artificial intelligenceMathematicsConjectureDiscrete cosine transformTRACE (psycholinguistics)Speech recognitionPattern recognition (psychology)LinguisticsPure mathematics

Abstract

fetched live from OpenAlex

Cosine similarity of contextual embeddings is used in many NLP tasks (e.g., QA, IR, MT) and metrics (e.g., BERTScore).Here, we uncover systematic ways in which word similarities estimated by cosine over BERT embeddings are understated and trace this effect to training data frequency.We find that relative to human judgements, cosine similarity underestimates the similarity of frequent words with other instances of the same word or other words across contexts, even after controlling for polysemy and other factors.We conjecture that this underestimation of similarity for high frequency words is due to differences in the representational geometry of high and low frequency words and provide a formal argument for the two-dimensional case.

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)0.032
metaresearch head score (Gemma)0.188
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: Methods · Consensus signal: Methods
Teacher disagreement score0.032
Threshold uncertainty score0.171

Distilled classifier scores by category (both heads)

CategoryCodexGemma
Metaresearch0.0320.188
Meta-epidemiology (narrow)0.0010.001
Meta-epidemiology (broad)0.0030.001
Bibliometrics0.0050.011
Science and technology studies0.0020.005
Scholarly communication0.0060.010
Open science0.0040.004
Research integrity0.0030.005
Insufficient payload (model declined to judge)0.0020.002

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 head0.252
Teacher spread0.225 · 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
GenreMethods

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

Citations54
Published2022
Admission routes1
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207