Pith. sign in

REVIEW 5 cited by

Multimodal Latent Language Modeling with Next-Token Diffusion

Not yet reviewed by Pith; the record is open.

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 2412.08635 v1 pith:CQQYATJY submitted 2024-12-11 cs.CL cs.CVcs.LG

classification cs.CLcs.CVcs.LG
keywords latentlmmultimodaldatamodelscontinuousdiffusiongenerationlanguage
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
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.

Discussion (0). Continue with ORCID to comment.

Forward citations

Cited by 5 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score. Full citation record

  1. FeynmanBench: Benchmarking Multimodal LLMs on Diagrammatic Physics Reasoning

    cs.AI 2026-04 unverdicted novelty 8.0 of 10

    FeynmanBench is the first benchmark for evaluating multimodal LLMs on diagrammatic reasoning with Feynman diagrams, revealing systematic failures in enforcing physical constraints and global topology.

  2. RobustSpeechFlow: Learning Robust Text-to-Speech Trajectories via Augmentation-based Contrastive Flow Matching

    cs.SD 2026-05 unverdicted novelty 6.0 of 10

    RobustSpeechFlow improves TTS alignment robustness by extending contrastive flow matching with length-preserving repeat and skip latent augmentations, lowering WER from 1.44 to 1.38 on Seed-TTS-eval and CER on ZERO500.

  3. Generative Audio Language Modeling with Continuous-valued Tokens and Masked Next-Token Prediction

    eess.AS 2025-07 conditional novelty 6.0 of 10

    A causal audio language model with continuous-valued tokens and masked next-token prediction matches diffusion-based text-to-audio quality with smaller, streamable models.

  4. ShareGPT-4o-Image: Aligning Multimodal Models with GPT-4o-Level Image Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    A 91K GPT-4o-generated image and editing dataset, and a fine-tuned open model Janus-4o, report improved text-to-image scores and new editing ability.

  5. MADFormer: Mixed Autoregressive and Diffusion Transformers for Continuous Image Generation

    cs.CV 2025-06 conditional novelty 4.0 of 10

    An empirical testbed shows that allocating more transformer layers to autoregressive conditioning improves FID under low inference budgets, while block size should scale with image resolution.

Pith tools