Related papers: Cross-Layer Discrete Concept Discovery for Interpr…

Understanding how generative models represent and transform data is a foundational problem in deep learning interpretability. While mechanistic interpretability of discriminative architectures has yielded substantial insights, relatively…

Variational Auto-Encoders (VAEs) are capable of learning latent representations for high dimensional data. However, due to the i.i.d. assumption, VAEs only optimize the singleton variational distributions and fail to account for the…

Deep generative models are reported to be useful in broad applications including image generation. Repeated inference between data space and latent space in these models can denoise cluttered images and improve the quality of inferred…

Recently, the cross-modal pre-training task has been a hotspot because of its wide application in various down-streaming researches including retrieval, captioning, question answering and so on. However, exiting methods adopt a one-stream…

Vector-quantized variational autoencoders (VQ-VAEs) are central to models that rely on high reconstruction fidelity, from neural compression to generative pipelines. Hierarchical extensions, such as VQ-VAE2, are often credited with superior…

Large Language Models (LLMs) have demonstrated remarkable capabilities across diverse natural language processing tasks. However, their extensive memory requirements, particularly due to KV cache growth during long-text understanding and…

Purpose: Handling heterogeneous and mixed data types has become increasingly critical with the exponential growth in real-world databases. While deep generative models attempt to merge diverse data views into a common latent space, they…

Large Language Models (LLMs) excel at reasoning and planning when trained on chainof-thought (CoT) data, where the step-by-step thought process is explicitly outlined by text tokens. However, this results in lengthy inputs where many words…

Mixture of Experts (MoE) architectures have recently advanced the scalability and adaptability of large language models (LLMs) for continual multimodal learning. However, efficiently extending these models to accommodate sequential tasks…

In recent years, the task of video prediction-forecasting future video given past video frames-has attracted attention in the research community. In this paper we propose a novel approach to this problem with Vector Quantized Variational…

Understanding the structure of complex, nonstationary, high-dimensional time-evolving signals is a central challenge in scientific data analysis. In many domains, such as speech and biomedical signal processing, the ability to learn…

The encoder-decoder dialog model is one of the most prominent methods used to build dialog systems in complex domains. Yet it is limited because it cannot output interpretable actions as in traditional systems, which hinders humans from…

For autoregressive (AR) modeling of high-resolution images, vector quantization (VQ) represents an image as a sequence of discrete codes. A short sequence length is important for an AR model to reduce its computational costs to consider…

Learning interpretable and disentangled representations of data is a key topic in machine learning research. Variational Autoencoder (VAE) is a scalable method for learning directed latent variable models of complex data. It employs a clear…

Vector Quantization (VQ) has become the cornerstone of tokenization for many multimodal Large Language Models and diffusion synthesis. However, existing VQ paradigms suffer from a fundamental conflict: they enforce discretization before the…

The integration of Vector Quantised Variational AutoEncoder (VQ-VAE) with autoregressive models as generation part has yielded high-quality results on image generation. However, the autoregressive models will strictly follow the progressive…

We introduce an information-theoretic framework that uses variational autoencoders (VAEs) to extract compact, physically interpretable manifolds from high-dimensional flow-field data. To this end, the Kullback--Leibler (KL) divergence in…

Text-to-Image generation in the general domain has long been an open problem, which requires both a powerful generative model and cross-modal understanding. We propose CogView, a 4-billion-parameter Transformer with VQ-VAE tokenizer to…

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…

Understanding what deep network models capture in their learned representations is a fundamental challenge in computer vision. We present a new methodology to understanding such vision models, the Visual Concept Connectome (VCC), which…