Multimodal Latent Language Modeling with Next-Token Diffusion

• URL: http://arxiv.org/abs/2412.08635v1
• Date: Wed, 11 Dec 2024 18:57:32 GMT
• Title: Multimodal Latent Language Modeling with Next-Token Diffusion
• Authors: Yutao Sun, Hangbo Bao, Wenhui Wang, Zhiliang Peng, Li Dong, Shaohan Huang, Jianyong Wang, Furu Wei,
• Abstract summary: Multimodal generative models require a unified approach to handle both discrete data (e.g., text and code) and continuous data (e.g., image, audio, video) We propose Latent Language Modeling (LatentLM), which seamlessly integrates continuous and discrete data using causal Transformers.
• Score: 111.93906046452125
• License:
• Abstract: Multimodal generative models require a unified approach to handle both discrete data (e.g., text and code) and continuous data (e.g., image, audio, video). In this work, we propose Latent Language Modeling (LatentLM), which seamlessly integrates continuous and discrete data using causal Transformers. Specifically, we employ a variational autoencoder (VAE) to represent continuous data as latent vectors and introduce next-token diffusion for autoregressive generation of these vectors. Additionally, we develop $\sigma$-VAE to address the challenges of variance collapse, which is crucial for autoregressive modeling. Extensive experiments demonstrate the effectiveness of LatentLM across various modalities. In image generation, LatentLM surpasses Diffusion Transformers in both performance and scalability. When integrated into multimodal large language models, LatentLM provides a general-purpose interface that unifies multimodal generation and understanding. Experimental results show that LatentLM achieves favorable performance compared to Transfusion and vector quantized models in the setting of scaling up training tokens. In text-to-speech synthesis, LatentLM outperforms the state-of-the-art VALL-E 2 model in speaker similarity and robustness, while requiring 10x fewer decoding steps. The results establish LatentLM as a highly effective and scalable approach to advance large multimodal models.

Related papers

• DiTAR: Diffusion Transformer Autoregressive Modeling for Speech Generation [24.85655658070008]
  Diffusion Transformer Autoregressive Modeling (DiTAR) is a patch-based autoregressive framework combining a language model with a diffusion transformer. In zero-shot speech generation, DiTAR achieves state-of-the-art performance in robustness, speaker similarity, and naturalness.
  arXiv  Detail & Related papers  (2025-02-06T10:09:49Z)
• Scalable Language Models with Posterior Inference of Latent Thought Vectors [52.63299874322121]
  Latent-Thought Language Models (LTMs) incorporate explicit latent thought vectors that follow an explicit prior model in latent space. LTMs possess additional scaling dimensions beyond traditional LLMs, yielding a structured design space. LTMs significantly outperform conventional autoregressive models and discrete diffusion models in validation perplexity and zero-shot language modeling.
  arXiv  Detail & Related papers  (2025-02-03T17:50:34Z)
• Dual Diffusion for Unified Image Generation and Understanding [32.7554623473768]
  We propose a large-scale and fully end-to-end diffusion model for multi-modal understanding and generation. We leverage a cross-modal maximum likelihood estimation framework that simultaneously trains the conditional likelihoods of both images and text jointly. Our model attained competitive performance compared to recent unified image understanding and generation models.
  arXiv  Detail & Related papers  (2024-12-31T05:49:00Z)
• ACDiT: Interpolating Autoregressive Conditional Modeling and Diffusion Transformer [95.80384464922147]
  Continuous visual generation requires the full-sequence diffusion-based approach. We present ACDiT, an Autoregressive blockwise Conditional Diffusion Transformer. We demonstrate that ACDiT can be seamlessly used in visual understanding tasks despite being trained on the diffusion objective.
  arXiv  Detail & Related papers  (2024-12-10T18:13:20Z)
• Diffusion Glancing Transformer for Parallel Sequence to Sequence Learning [52.72369034247396]
  We propose the diffusion glancing transformer, which employs a modality diffusion process and residual glancing sampling. DIFFGLAT achieves better generation accuracy while maintaining fast decoding speed compared with both autoregressive and non-autoregressive models.
  arXiv  Detail & Related papers  (2022-12-20T13:36:25Z)
• Unified Discrete Diffusion for Simultaneous Vision-Language Generation [78.21352271140472]
  We present a unified multimodal generation model that can conduct both the "modality translation" and "multi-modality generation" tasks. Specifically, we unify the discrete diffusion process for multimodal signals by proposing a unified transition matrix. Our proposed method can perform comparably to the state-of-the-art solutions in various generation tasks.
  arXiv  Detail & Related papers  (2022-11-27T14:46:01Z)
• Improve Variational Autoencoder for Text Generationwith Discrete Latent Bottleneck [52.08901549360262]
  Variational autoencoders (VAEs) are essential tools in end-to-end representation learning. VAEs tend to ignore latent variables with a strong auto-regressive decoder. We propose a principled approach to enforce an implicit latent feature matching in a more compact latent space.
  arXiv  Detail & Related papers  (2020-04-22T14:41:37Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.