相关论文: ELSA: Enhanced Local Self-Attention for Vision Tra…

Current state-of-the-art semantic role labeling (SRL) uses a deep neural network with no explicit linguistic features. However, prior work has shown that gold syntax trees can dramatically improve SRL decoding, suggesting the possibility of…

Reducing the key-value (KV) cache size is a crucial step toward enabling efficient inference in large language models (LLMs), especially under latency and memory constraints. While Multi-Head Attention (MHA) offers strong representational…

Learning new classes without forgetting is crucial for real-world applications for a classification model. Vision Transformers (ViT) recently achieve remarkable performance in Class Incremental Learning (CIL). Previous works mainly focus on…

Key-Value (KV) cache memory and bandwidth increasingly dominate large language model inference cost in long-context and long-generation regimes. Architectures such as multi-head latent attention (MLA) and hybrid sliding-window attention…

Multi-head Latent Attention (MLA) is an innovative architecture proposed by DeepSeek, designed to ensure efficient and economical inference by significantly compressing the Key-Value (KV) cache into a latent vector. Compared to MLA,…

Recent advances in transformer-based Large Language Models (LLMs) have demonstrated remarkable capabilities across various tasks. However, their quadratic computational complexity concerning sequence length remains a significant bottleneck…

Modern sequence modeling is dominated by two families: Transformers, whose self-attention can access arbitrary elements of the visible sequence, and structured state-space models, which propagate information through an explicit recurrent…

Optimizing vision models purely for classification accuracy can impose an alignment tax, degrading human-like scanpaths and limiting interpretability. We introduce EVA, a neuroscience-inspired hard-attention mechanistic testbed that makes…

Humans can picture a sound scene given an imprecise natural language description. For example, it is easy to imagine an acoustic environment given a phrase like "the lion roar came from right behind me!". For a machine to have the same…

Window-based transformers have demonstrated outstanding performance in super-resolution tasks due to their adaptive modeling capabilities through local self-attention (SA). However, they exhibit higher computational complexity and inference…

Despite great success across various multimodal tasks, Large Vision-Language Models (LVLMs) often encounter object hallucinations with generated textual responses being inconsistent with the actual objects in images. We examine different…

Vision Transformers (ViT) have shown their competitive advantages performance-wise compared to convolutional neural networks (CNNs) though they often come with high computational costs. To this end, previous methods explore different…

Transformers have recently shown superior performances on various vision tasks. The large, sometimes even global, receptive field endows Transformer models with higher representation power over their CNN counterparts. Nevertheless, simply…

Recent Few-Shot Learning (FSL) methods put emphasis on generating a discriminative embedding features to precisely measure the similarity between support and query sets. Current CNN-based cross-attention approaches generate discriminative…

Convolutions (Convs) and multi-head self-attentions (MHSAs) are typically considered alternatives to each other for building vision backbones. Although some works try to integrate both, they apply the two operators simultaneously at the…

The hallucination problem in multimodal large language models (MLLMs) remains a common issue. Although image tokens occupy a majority of the input sequence of MLLMs, there is limited research to explore the relationship between image tokens…

In this paper,an Enhanced Self-Attention (ESA) mechanism has been put forward for robust feature extraction.The proposed ESA is integrated with the recursive gated convolution and self-attention mechanism.In particular, the former is used…

Although Large Vision-Language Models (LVLMs) have made substantial progress, hallucination, where generated text is not grounded in the visual input, remains a challenge. As LVLMs become stronger, previously reported hallucination…

Cross-Domain Few-Shot Learning (CDFSL) aims to adapt large-scale pretrained models to specialized target domains with limited samples, yet the few-shot fine-tuning of vision-language models like CLIP remains underexplored. By establishing…

Balancing accuracy and latency on high-resolution images is a critical challenge for lightweight models, particularly for Transformer-based architectures that often suffer from excessive latency. To address this issue, we introduce…