In this study, we investigate the task of data pre-selection, which aims to select instances for labeling from an unlabeled dataset through a single pass, thereby optimizing performance for undefined downstream tasks with a limited annotation budget. Previous approaches to data pre-selection relied solely on visual features extracted from foundation models, such as CLIP and BLIP-2, but largely ignored the powerfulness of text features. In this work, we argue that, with proper design, the joint feature space of both vision and text can yield a better representation for data pre-selection. To this end, we introduce UP-DP, a simple yet effective unsupervised prompt learning approach that adapts vision-language models, like BLIP-2, for data pre-selection. Specifically, with the BLIP-2 parameters frozen, we train text prompts to extract the joint features with improved representation, ensuring a diverse cluster structure that covers the entire dataset. We extensively compare our method with the state-of-the-art using seven benchmark datasets in different settings, achieving up to a performance gain of 20%. Interestingly, the prompts learned from one dataset demonstrate significant generalizability and can be applied directly to enhance the feature extraction of BLIP-2 from other datasets. To the best of our knowledge, UP-DP is the first work to incorporate unsupervised prompt learning in a vision-language model for data pre-selection.

Let qu(θ) set as Eq. For notational simplicity, let θ0 = θ(t 1). B.1 Hyperparameter settings Training In Table 2, we enumerate the hyperparameters used for our results in Section 5. Since we use expert trajectories for all methods to train the Bayesian pseudocoresets, we refer to hyperparameters related to expert trajectories, such as the number of SGD steps or the maximum random starting points, described in [8]. We found that a slightly shorter expert training step is better for BPC-fKL, so we used an expert step 1 epoch shorter than BPC-W. For each setting, we used the best learning rate from a hyperparameter sweep over {0.01,0.02,0.03,0.04}.

Neuron interpretation offers valuable insights into how knowledge is structured within a deep neural network model.

Hence, the above formulation of set function is submodular and is an instance of concave over modular function. Inour setting, we use labeled target dataDt asthe validation set. The completed-SNE loss is defined as a combination of Land cross-entropy loss on source and targetdomain. Table 8 shows the training times for this setting. Again, we see that all instantiationsofORIENTachieve 2.5 speed-upcomparedtoFull.

We design a suite of benchmarks to evaluate the quality of the generated MILP instances.

We design a suite of benchmarks to evaluate the quality of the generated MILP instances.

Seed value of 42 is used in generator argument in random_split function of torch.