AI Paper Atlas

Research map / Vision-language and multimodal / Human and embodied modeling

Domino: Discovering Systematic Errors with Cross-Modal Embeddings

Sabri Eyuboglu, Maya Varma, Khaled Kamal Saab, Jean-Benoit Delbrouck, Christopher Lee-Messer, Jared Dunnmon, James Zou, Christopher Re

ICLR 2022 · Vision-language and multimodal › Human and embodied modeling · approach: Scientific domain vision datasets · 38 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 cross-modal CLIP embeddings instead of BiT embeddings (ablation)Domino on natural images with synthetic models · Natural imagesmean precision-at-10: a 9 percentage point increase over BiT embeddingsImproved
Use cross-modal CLIP embeddings instead of random activations (ablation)Domino on natural images with synthetic models · Natural imagesmean precision-at-10: a 23 percentage point increase over random activationsImproved
Use CLIP embeddings instead of BiT embeddings (ablation)Domino on natural images with trained models · Natural imagesmean precision-at-10No clear effect
Use CLIP embeddings instead of trained-classifier activations (ablation)Domino on natural images with trained models · Natural imagesmean precision-at-10: nearly 15 percentage pointsImproved
Use ConVIRT embeddings instead of BiT embeddings (ablation)Domino on medical images with synthetic models · Medical imagesmean precision-at-10: a 7 point improvement over the best unimodal embeddings (BiT)Improved
Use cross-modal embeddings instead of unimodal embeddings (ablation)Domino on medical time-series with synthetic models · Medical time-seriesmean precision-at-10: a 17 point gain over using unimodal embeddings 0.532Improved
Use the error-aware mixture model instead of George (ablation)Domino with cross-modal embeddings on natural-image noisy and rare slices · Natural images, noisy and rare slicesmean precision-at-10: a 105% improvement over the next-best method, GeorgeImproved
Use the error-aware mixture model instead of ConfusionSDM (ablation)Domino with cross-modal embeddings on correlation slices · Natural images, correlation slicesslice recovery performanceHurt
Generate natural-language slice descriptions from cross-modal embeddingsDomino on natural-image slice settings · Natural images, slice descriptionssettings with exact slice name ranked first: 34.7% of rare slices, 41.0% of correlation slices, and 39.0% of noisy label slicesImproved

Later papers that compared with this method

1 later accepted paper reports Domino as a baseline in their results tables; 1 of them report a better number on most of the results they share with it.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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