MétaCan
Menu
Back to cohort
Record W4319986073 · doi:10.17760/d20467253

Self-repetition in abstractive neural summarizers

2022· article· en· W4319986073 on OpenAlexaff
Nikita Salkar

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicTopic Modeling
Canadian institutionsArtificial Intelligence in Medicine (Canada)
FundersU.S. National Library of MedicineNational Institutes of HealthNational Science Foundation
KeywordsRepetition (rhetorical device)Computer scienceNatural language processingArtificial intelligenceQualitative analysisLinguisticsQualitative researchSociology

Abstract

fetched live from OpenAlex

We provide a quantitative and qualitative analysis of self-repetition in the output of neural summarizers. We measure self-repetition as the number of n-grams of length four or longer that appear in multiple outputs of the same system. We analyze the behavior of three popular summarization architectures (BART, T5, and Pegasus), fine-tuned on five representative datasets. In a regression analysis, we find that the three architectures have different propensities for repeating con-tent in their summaries for different inputs, with BART being the most and considerably prone to self-repeat. Fine-tuning on more abstractive data is associated with a higher rate of self-repetition. In the qualitative analysis, we find systems produce artefacts such as ads and disclaimers unrelated to the content being summarized, as well as formulaic phrases common in the fine-tuning domain, which often are hallucinations not supported by the original input. While some hallucinations are suitable in the context of the summary and its input, often times they are incongruous and completely unrelated to the input. Our approach to corpus-level analysis of self-repetition may help practitioners clean up training data for summarizers, and ultimately support methods for minimizing the amount of hallucinations in abstractive neural summaries.--Author's abstract

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 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)0.000
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: Simulation or modeling · Consensus signal: none
GenreCandidate signal: Empirical · Consensus signal: Empirical
Teacher disagreement score0.798
Threshold uncertainty score0.175

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
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 science0.0000.000
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 head0.014
GPT teacher head0.228
Teacher spread0.214 · 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 teacher head, 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

Citations3
Published2022
Admission routes1
Has abstractyes

Explore more

Same topicTopic ModelingFrench-language works237,207