Related papers: SALSA: Semantically-Aware Latent Space Autoencoder

Pseudo-label learning is widely used in semantic segmentation, particularly in label-scarce scenarios such as unsupervised domain adaptation (UDA) and semisupervised learning (SSL). Despite its success, this paradigm can generate erroneous…

Despite significant progress in Unified Multimodal Retrieval (UMR) powered by Large Multimodal Models (LMMs), existing embedding methods primarily focus on sample-level objectives via contrastive learning while overlooking the crucial…

Our work introduces SAVeD (Semantically Aware Version Detection), a contrastive learning-based framework for identifying versions of structured datasets without relying on metadata, labels, or integration-based assumptions. SAVeD addresses…

At present, deep neural network methods have played a dominant role in face alignment field. However, they generally use predefined network structures to predict landmarks, which tends to learn general features and leads to mediocre…

Sparse autoencoders are a promising new approach for decomposing language model activations for interpretation and control. They have been applied successfully to vision transformer image encoders and to small-scale diffusion models.…

The manifold hypothesis states that high-dimensional data can be modeled as lying on or near a low-dimensional, nonlinear manifold. Variational Autoencoders (VAEs) approximate this manifold by learning mappings from low-dimensional latent…

Zero-shot skeleton-based action recognition (ZSAR) aims to recognize action classes without any training skeletons from those classes, relying instead on auxiliary semantics from text. Existing approaches frequently depend on explicit…

Deep neural networks often under-perform on tabular data due to their sensitivity to irrelevant features and a spectral bias toward smooth, low-frequency functions. These limitations hinder their ability to capture the sharp, high-frequency…

Semantic segmentation on LiDAR imaging is increasingly gaining attention, as it can provide useful knowledge for perception systems and potential for autonomous driving. However, collecting and labeling real LiDAR data is an expensive and…

Zero-Shot Learning (ZSL) is typically achieved by resorting to a class semantic embedding space to transfer the knowledge from the seen classes to unseen ones. Capturing the common semantic characteristics between the visual modality and…

When given two similar images, humans identify their differences by comparing the appearance (e.g., color, texture) with the help of semantics (e.g., objects, relations). However, mainstream binary change detection models adopt a supervised…

Sparse autoencoders (SAEs) are an unsupervised method for learning a sparse decomposition of a neural network's latent representations into seemingly interpretable features. Despite recent excitement about their potential, research…

We explore a framework for protein sequence representation learning that decomposes the task between manifold learning and distributional modelling. Specifically we present a Latent Space Diffusion architecture which combines a protein…

Decomposing model activations into interpretable components is a key open problem in mechanistic interpretability. Sparse autoencoders (SAEs) are a popular method for decomposing the internal activations of trained transformers into sparse,…

While different neural models often exhibit latent spaces that are alike when exposed to semantically related data, this intrinsic similarity is not always immediately discernible. Towards a better understanding of this phenomenon, our work…

Sparse Autoencoders (SAEs) have emerged as a powerful tool for interpreting the internal representations of CLIP vision encoders, yet existing analyses largely focus on the semantic meaning of individual features. We introduce information…

Vanilla pixel-level classifiers for semantic segmentation are based on a certain paradigm, involving the inner product of fixed prototypes obtained from the training set and pixel features in the test image. This approach, however,…

Unsupervised sentence embeddings task aims to convert sentences to semantic vector representations. Most previous works directly use the sentence representations derived from pretrained language models. However, due to the token bias in…

Training recognition models with synthetic images have achieved remarkable results in text recognition. However, recognizing text from real-world images still faces challenges due to the domain shift between synthetic and real-world text…

Transformer models have become state-of-the-art in decoding stimuli and behavior from neural activity, significantly advancing neuroscience research. Yet greater transparency in their decision-making processes would substantially enhance…