Predictive Data Attribution
An extension of data attribution that introduces a surrogate model to predict the performance change caused by adding a given data point to the training set, without actually training the target policy, thereby avoiding the high cost of repeated retraining.
Timeline (last 90 days)
GUILD introduces a data evaluation stage that uses predictive data attribution to approximately estimate the impact of collected trajectories on policy performance in the target deployment scenario.
The accuracy of predictive data attribution is constrained by the quality of the surrogate model, which is a key prerequisite for the reliability of GUILD's feedback signal.
GUILD currently still depends on the quality of the VLM's semantic priors, and the accuracy of predictive attribution affects the effectiveness of the feedback signal.
All Facts (3)
GUILD introduces a data evaluation stage that uses predictive data attribution to approximately estimate the impact of collected trajectories on policy performance in the target deployment scenario.
50%UnverifiedThe accuracy of predictive data attribution is constrained by the quality of the surrogate model, which is a key prerequisite for the reliability of GUILD's feedback signal.
50%UnverifiedGUILD currently still depends on the quality of the VLM's semantic priors, and the accuracy of predictive attribution affects the effectiveness of the feedback signal.
50%