Labelbox•April 8, 2021
Labelbox lets you manage every aspect of creating high quality training data for your ML initiatives. Today, we’re excited to make collaboration and communication within and across teams in Labelbox even easier with the introduction of Issues and Comments. Labelers, reviewers, and other teammates involved in creating and reviewing training data can now start conversation threads in the image editor while labeling or reviewing labels. By centralizing and streamlining communication within Labelbox, teams can deliver higher quality training data and focus on getting to production faster. Get started in the app or by reading our documentation.
Labelers can face uncertainty while labeling, especially when they’re onboarding or working with particularly complex or subjective data. Issues and Comments creates a simple and reliable channel for escalating questions to reviewers or subject matter experts. A labeler can create an issue to ask a question, submit the label completed to the best of their ability, and receive feedback and clarification during the label review process.
Review is a critical part of QA/QC and an important opportunity for providing feedback and coaching to labelers. Reviewers can now create issues and add comments to provide pinpoint feedback when sending labels back to the labeling queue for correction. Labelers and reviewers can contribute additional comments to the conversation until the issue is marked as resolved. This is the fastest, most direct, and most detailed way to provide feedback on a label, and is much more reliable than routing feedback through chat, email, or other channels outside of the Labelbox platform.
Labelers, reviewers, and admins alike can access the new “Issues” tab within a project to see every issue they’re a part of in one place, along with who most recently commented and when. Labelers can create, comment on, and resolve any issue on a label they created while project reviewers and admins have access to issues on any label in their project. Teams can now rely on the Issues tab to show them the most recent snapshot of all conversations related to the labels in their project. With a consolidated view of this activity, teams can resolve labeling issues faster than ever.
Labelbox is a flexible, powerful platform for bringing together internal and external stakeholders to create high-quality training data. Issues and Comments is just another step in simplifying the complexities of managing diverse teams. Other improvements and advantages include:
Simplified user management: We've added "Pending Invitations'' to project-level and organization-level members tabs to list any users who haven't yet accepted an invitation. Plus, you can now resend or revoke pending invitations.
Role-based access control: When adding members, you can specify their role and permissions at either the project or organization level. This way, whether you’re adding labelers, reviewers, team managers, or admins, you can ensure they have the right permissions to the right projects.
Issues and Comments gives teams a channel to ensure that labels are of a high and consistent quality. We’re excited to bring teams closer together and help make communication and collaboration easier than ever.
Labelbox•September 26, 2022
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.
Labelbox•September 26, 2022
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.