Related papers: Visual Words Meet BM25: Sparse Auto-Encoder Visual…

Medical image segmentation, or computing voxelwise semantic masks, is a fundamental yet challenging task to compute a voxel-level semantic mask. To increase the ability of encoder-decoder neural networks to perform this task across large…

The past few years have witnessed increasing interests in applying deep learning to video compression. However, the existing approaches compress a video frame with only a few number of reference frames, which limits their ability to fully…

Classification of partially occluded images is a highly challenging computer vision problem even for the cutting edge deep learning technologies. To achieve a robust image classification for occluded images, this paper proposes a novel…

We present a real-time visual-inertial dense mapping method capable of performing incremental 3D mesh reconstruction with high quality using only sequential monocular images and inertial measurement unit (IMU) readings. 6-DoF camera poses…

Sparse auto-encoders (SAEs) have re-emerged as a prominent method for mechanistic interpretability, yet they face two significant challenges: the non-smoothness of the $L_1$ penalty, which hinders reconstruction and scalability, and a lack…

DeepSeek-OCR shows that rendered text can be reconstructed from a small number of vision tokens, sparking excitement about using vision as a compression medium for long textual contexts. But this pipeline requires rendering token embeddings…

We present the first systematic study of Sparse Autoencoders (SAEs) on video representations. Standard SAEs decompose video into interpretable, monosemantic features but destroy temporal coherence: hard TopK selection produces unstable…

Neural implicit representations have revolutionized dense multi-view surface reconstruction, yet their performance significantly diminishes with sparse input views. A few pioneering works have sought to tackle the challenge of sparse-view…

In retrieval-augmented coding, failures often begin when the relevant file is absent from the retrieved context. Under frozen generic tokenization, where a BM25 index has been built by a search system whose analyzer the practitioner does…

Current pre-trained language model approaches to information retrieval can be broadly divided into two categories: sparse retrievers (to which belong also non-neural approaches such as bag-of-words methods, e.g., BM25) and dense retrievers.…

As the first step in automated natural language processing, representing words and sentences is of central importance and has attracted significant research attention. Different approaches, from the early one-hot and bag-of-words…

Detecting illicit visual content demands more than image-level NSFW flags; moderators must also know what objects make an image illegal and where those objects occur. We introduce a zero-shot pipeline that simultaneously (i) detects if an…

Disentangling model activations into meaningful features is a central problem in interpretability. However, the absence of ground-truth for these features in realistic scenarios makes validating recent approaches, such as sparse dictionary…

In this paper, we propose a new self-supervised method, which is called Denoising Masked AutoEncoders (DMAE), for learning certified robust classifiers of images. In DMAE, we corrupt each image by adding Gaussian noises to each pixel value…

Dictionary learning can be used for image superresolution by learning a pair of coupled dictionaries of image patches from high-resolution and low-resolution image pairs such that the corresponding pairs share the same sparse vector when…

We introduce a method to learn unsupervised semantic visual information based on the premise that complex events can be decomposed into simpler events and that these simple events are shared across several complex events. We first employ a…

Vision transformers in vision-language models typically use the same amount of compute for every image, regardless of whether it is simple or complex. We propose ICAR (Image Complexity-Aware Retrieval), an adaptive computation approach that…

Sparse Autoencoders (SAEs) are increasingly used to interpret foundation models, but their role as an actionable intervention space remains less understood, especially in vision. We study whether sparse visual features can be used not only…

Recent vision-language models (VLMs) typically rely on a single vision encoder trained with contrastive image-text objectives, such as CLIP-style pretraining. While contrastive encoders are effective for cross-modal alignment and retrieval,…

Sparse autoencoders (SAEs) provide a powerful mechanism for decomposing the dense representations produced by Large Language Models (LLMs) into interpretable latent features. We posit that SAEs constitute a natural foundation for Learned…