AI Paper Atlas

Research map / Probabilistic and scientific modeling / Molecules and proteins

Personalized Benchmarking with the Ludwig Benchmarking Toolkit

Avanika Narayan, Piero Molino, Karan Goel, Willie Neiswanger, Christopher Re

NeurIPS 2021 · Probabilistic and scientific modeling › Molecules and proteins · approach: Statistical analysis of ML practice · 1 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
Use DistilBERT rather than other evaluated modelsText classification models · Text classification across 9 datasetsinference efficiency-performance tradeoffImproved
Use BERT rather than RoBERTa or T5-smallText classification models · Text classification across datasetsinference latencyImproved
Increase model size / use larger modelsText classification models · Text classification across 9 datasetsaccuracyImproved
Use pretrained models rather than models trained from scratchText classification models · Text classification datasetsaccuracy varianceMixed
Evaluate on midsize rather than small datasetsModels compared across dataset sizes · Text classification datasetsvariance in model performanceImproved
Increase average sentence lengthText classification datasets · Text classification datasetsaccuracyImproved
Fine-tune BERTBERT models · TextAttack adversarial attacks across datasetsattack success rateHurt
Use RoBERTa instead of BERTPretrained text classification models · TextAttack adversarial attacks across datasetsattack success rateImproved
Use RNN instead of BERTText classification models · TextAttack adversarial attacks across datasetsattack success rateImproved
Use pretrained rather than scratch-trained modelsText classification models · Model convergenceepochs until best checkpointMixed

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy