Open-vocabulary object detection has benefited greatly from pretrained vision-language
models, but is still limited by the amount of available detection training data. While detection …

Since the introduction of the Vision Transformer (ViT), researchers have sought to make ViTs
more efficient by removing redundant information in the processed tokens. While different …

Building on the foundations of language modeling in natural language processing, Next
Token Prediction (NTP) has evolved into a versatile training objective for machine learning …

The performance of transformers for time-series forecasting has improved significantly.
Recent architectures learn complex temporal patterns by segmenting a time-series into …

Abstract We present Agglomerative Token Clustering (ATC), a novel token merging method
that consistently outperforms previous token merging and pruning methods across image …

Recent token reduction methods for Vision Transformers (ViTs) incorporate token merging,
which measures the similarities between token embeddings and combines the most similar …

The accurate segmentation of medical images is crucial for diagnosing and treating
diseases. Recent studies demonstrate that vision transformer-based methods have …

Increasing the throughput of the Transformer architecture, a foundational component used in
numerous state-of-the-art models for vision and language tasks (eg, GPT, LLaVa), is an …

Token compression expedites the training and inference of Vision Transformers (ViTs) by
reducing the number of the redundant tokens, eg, pruning inattentive tokens or merging …

Time series forecasting has attracted significant attention in recent decades. Previous
studies have demonstrated that the Channel-Independent (CI) strategy improves forecasting …