Blog

Find model failures with auto-generated metrics

New in Labelbox Model: Teams can soon surface high-impact model failures with auto-generated metrics.

Optimize pre-labeling efficiency with automation metrics

New in Labelbox Annotate: Optimize model-assisted labeling, queue & review better, and more.

Leverage powerful embeddings to surface high-impact data in one click

New in Labelbox Catalog: One-click similarity search, faster data row identification, and more.

Measure and improve pre-labeling workflows with the automation efficiency score

Most ML teams operate blindly or struggle to quantify the efficiency of their pre-labeling process. Today, Labelbox is launching the automation efficiency score –a new way to track, measure, and improve your pre-labeling efficiency and time savings in single dashboard.

Better understand and visualize data rows to surface high-impact data faster

Labelbox has made visibility and usability enhancements across Catalog, Annotate, and Model. Teams can now identify and fix data row processing errors, easily reorder ontology nodes, better visualize ground truth and predictions in Model, and more.

Annotate documents, HTML files, high-resolution pathology slides, and queue & review data better with custom workflows

Leverage our document editor for both NER and OCR annotations, view data in a specific manner with our HTML editor, deep zoom on tiled imagery for medical use cases, and more. As we continue to release Workflows across our customer base, teams can more effectively queue and review their data rows.

Prevent duplicate data uploads, save data slices, metadata metrics, and more

We've introduced a new system to identify data rows to allow teams to query faster and prevent duplicate uploads, introduced metadata analytics, and are making a host of exciting improvements to our embedding capabilities.

Better visualize and track model errors at every iteration

Find and fix label errors for text data, understand model weaknesses with in-depth metrics, and easily compare results for different model runs with different hyperparameters.

Accelerate AI development and search for more granular data with new features

We’ve focused recent product development on a host of improvements to help drive greater insight and productivity across annotation, data curation, and model training workflows.

Labelbox joins Air Force’s $950 million JADC2 effort

Labelbox today announced it is one of the companies who have been awarded a $950 million ceiling indefinite-delivery/indefinite-quantity contract by the U.S. Air Force.