相关论文: Token Bottleneck: One Token to Remember Dynamics

Tokenization in video models, typically through patchification, generates an excessive and redundant number of tokens. This severely limits video efficiency and scalability. While recent trajectory-based tokenizers offer a promising…

In recent years, graph representation learning has gained significant popularity, which aims to generate node embeddings that capture features of graphs. One of the methods to achieve this is employing a technique called random walks that…

We propose a novel approach to performing fine-grained 3D manipulation of image content via a convolutional neural network, which we call the Transformable Bottleneck Network (TBN). It applies given spatial transformations directly to a…

Character-level language models obviate the need for separately trained tokenizers, but efficiency suffers from longer sequence lengths. Learning to combine character representations into tokens has made training these models more…

This work presents VTok, a unified video tokenization framework that can be used for both generation and understanding tasks. Unlike the leading vision-language systems that tokenize videos through a naive frame-sampling strategy, we…

Current vision models typically maintain a fixed correspondence between their representation structure and image space. Each layer comprises a set of tokens arranged "on-the-grid," which biases patches or tokens to encode information at a…

We present an approach to pose object recognition as next token prediction. The idea is to apply a language decoder that auto-regressively predicts the text tokens from image embeddings to form labels. To ground this prediction process in…

Concept Bottleneck Models (CBMs) offer inherent interpretability by initially translating images into human-comprehensible concepts, followed by a linear combination of these concepts for classification. However, the annotation of concepts…

Modeling visual data as tokens (i.e., image patches) using attention mechanisms, feed-forward networks or convolutions has been highly effective in recent years. Such methods usually have a common pipeline: a tokenization method, followed…

We present TokenDial, a framework for continuous, slider-style attribute control in pretrained text-to-video generation models. While modern generators produce strong holistic videos, they offer limited control over how much an attribute…

The analysis of events in dynamic environments poses a fundamental challenge in the development of intelligent agents and robots capable of interacting with humans. Current approaches predominantly utilize visual models. However, these…

Recent perception-generalist approaches based on language models have achieved state-of-the-art results across diverse tasks, including 3D scene layout estimation and 3D object detection, via unified architecture and interface. However,…

We introduce a new family of toy problems that combine features of linear-regression-style continuous in-context learning (ICL) with discrete associative recall. We pretrain transformer models on sample traces from this toy, specifically…

Image tokenizers form the foundation of modern text-to-image generative models but are notoriously difficult to train. Furthermore, most existing text-to-image models rely on large-scale, high-quality private datasets, making them…

The selective visual attention mechanism in the human visual system (HVS) restricts the amount of information to reach visual awareness for perceiving natural scenes, allowing near real-time information processing with limited computational…

With the growing adoption of vision-language-action models and world models in autonomous driving systems, scalable image tokenization becomes crucial as the interface for the visual modality. However, most existing tokenizers are designed…

Variational autoencoders (VAEs) are essential tools in end-to-end representation learning. However, the sequential text generation common pitfall with VAEs is that the model tends to ignore latent variables with a strong auto-regressive…

The event camera's low power consumption and ability to capture microsecond brightness changes make it attractive for various computer vision tasks. Existing event representation methods typically convert events into frames, voxel grids, or…

In this paper, we propose a novel token selective attention approach, ToSA, which can identify tokens that need to be attended as well as those that can skip a transformer layer. More specifically, a token selector parses the current…

Neural implicit mapping has emerged as a powerful paradigm for robotic navigation and scene understanding. However, real-world robotic deployment requires continual adaptation to changing environments under strict memory and computation…