MétaCan
Menu
Back to cohort
Record W4402671693 · doi:10.18653/v1/2024.acl-long.680

To Distill or Not to Distill? On the Robustness of Robust Knowledge Distillation

2024· article· en· W4402671693 on OpenAlex

Why this work is in the frame

A frame that forgets how it found something cannot be audited. These are the routes that admitted this work.

fundA Canadian funder is recorded on the work.
no affNo Canadian affiliation: this work is invisible to an affiliation-only frame.
No Canadian affiliation. An affiliation-only frame, the usual design, would never have seen this work. It is one of the works that make the case for inverting the frame.

Bibliographic record

Venuenot available
Typearticle
Languageen
FieldComputer Science
TopicBayesian Modeling and Causal Inference
Canadian institutionsnot available
FundersAlliance de recherche numérique du CanadaSocial Sciences and Humanities Research Council of CanadaNatural Sciences and Engineering Research Council of CanadaCanada Research Chairs
KeywordsRobustness (evolution)Computer scienceDistillationArtificial intelligenceMachine learningChromatographyBiologyChemistry

Abstract

fetched live from OpenAlex

Arabic is known to present unique challenges for Automatic Speech Recognition (ASR).On one hand, its rich linguistic diversity and wide range of dialects complicate the development of robust, inclusive models.On the other, current multilingual ASR models are compute-intensive and lack proper comprehensive evaluations.In light of these challenges, we distill knowledge from large teacher models into smaller student variants that are more efficient.We also introduce a novel human-annotated dataset covering five under-represented Arabic dialects for evaluation.We further evaluate both our models and existing SoTA multilingual models on both standard available benchmarks and our new dialectal data.Our best-distilled model's overall performance (45.0%WER) surpasses that of a SoTA model twice its size (SeamlessM4T-large-v2, WER=47.0%) and its teacher model (Whisper-large-v2, WER=55.1%), and its average performance on our new dialectal data (56.9%WER) outperforms all other models.To gain more insight into the poor performance of these models on dialectal data, we conduct an error analysis and report the main types of errors the different models tend to make.The GitHub repository for the project is available at https: //github.com/UBC-NLP/distill-whisper-ar.

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.

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: none
Teacher disagreement score0.964
Threshold uncertainty score0.390

Codex and Gemma teacher scores by category

CategoryCodexGemma
Metaresearch0.0000.000
Meta-epidemiology (narrow)0.0000.000
Meta-epidemiology (broad)0.0000.000
Bibliometrics0.0000.001
Science and technology studies0.0000.000
Scholarly communication0.0000.000
Open science0.0010.000
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 head0.083
GPT teacher head0.320
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

Quick stats

Citations3
Published2024
Admission routes1
Has abstractyes

Explore more

Same topicBayesian Modeling and Causal InferenceFrench-language works237,207