DataMecka’s human demonstrations test what robot models can learn before touching a robot
Capture hardware and reconstruction models turn human activity into candidate training data. The hard part is preserving what transfers to a different body.
Data
The datasets, collection systems and curation methods that determine what a model can learn.
Datasets · Collection & data rigs · Synthetic data · Curation & labeling
5 articles
Data
DataTwelveLabs targets the gap between human video and usable training data
Pegasus 1.6 brings first-person footage into a structured labeling pipeline. The test is whether it saves effort without teaching models the wrong actions.
DataSnorkel’s training environments turn expert judgment into a model curriculum
For models that take actions, a dataset increasingly needs states, tools and verifiable outcomes. The design of that environment determines what the model learns.
DataDatology puts the training budget inside the dataset
An October launch makes a consequential claim: companies may need better-selected tokens before they need a bigger model.
DataNVIDIA turns synthetic training data into a specification to inspect
NeMo Data Designer’s recent research and software updates move attention from generating examples to controlling what those examples teach.
No articles in this topic yet
Explore the other topics in this desk.