Pith. sign in

REVIEW 11 cited by

Liquid: Language Models are Scalable and Unified Multi-modal Generators

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

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 2412.04332 v4 pith:ZNGDDYI7 submitted 2024-12-05 cs.CV

Junfeng Wu , Yi Jiang , Chuofan Ma , Yuliang Liu , Hengshuang Zhao , Zehuan Yuan , Song Bai , Xiang Bai This is my paper
classification cs.CV
keywords liquidlanguagegenerationmodelsvisualllmsmodelmultimodal
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

We present Liquid, an auto-regressive generation paradigm that seamlessly integrates visual comprehension and generation by tokenizing images into discrete codes and learning these code embeddings alongside text tokens within a shared feature space for both vision and language. Unlike previous multimodal large language model (MLLM), Liquid achieves this integration using a single large language model (LLM), eliminating the need for external pretrained visual embeddings such as CLIP. For the first time, Liquid uncovers a scaling law that performance drop unavoidably brought by the unified training of visual and language tasks diminishes as the model size increases. Furthermore, the unified token space enables visual generation and comprehension tasks to mutually enhance each other, effectively removing the typical interference seen in earlier models. We show that existing LLMs can serve as strong foundations for Liquid, saving 100x in training costs while outperforming Chameleon in multimodal capabilities and maintaining language performance comparable to mainstream LLMs like LLAMA2. Liquid also outperforms models like SD v2.1 and SD-XL (FID of 5.47 on MJHQ-30K), excelling in both vision-language and text-only tasks. This work demonstrates that LLMs such as Qwen2.5 and GEMMA2 are powerful multimodal generators, offering a scalable solution for enhancing both vision-language understanding and generation. The code and models will be released at https://github.com/FoundationVision/Liquid.

Discussion (0). Sign in to comment.

Forward citations

Cited by 11 Pith papers

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

  1. Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes

    cs.CV 2026-08 conditional novelty 6.0 of 10

    Multimodal pretraining transfers asymmetrically: language boosts vision, understanding boosts generation, generation is mostly neutral, and early unified training prevents vision laziness.

  2. Twins: Learn to Predict Unified Representations with Focal Loss

    cs.CV 2026-07 conditional novelty 6.0 of 10

    Channel-wise concatenation of SigLIP2 and Flux VAE features into one token, trained with a focal-style flow-matching loss, yields a unified representation with 1.59 gFID on ImageNet 256 and VAE-level reconstruction.

  3. Unified Audio Intelligence Without Regressing on Text Intelligence

    cs.CL 2026-07 conditional novelty 6.0 of 10

    A unified 30B MoE audio-text LLM achieves state-of-the-art audio understanding, generation, and speech tasks while preserving text reasoning comparable to its text-only backbone.

  4. AutoNeural: Co-Designing Vision-Language Models for NPU Inference

    cs.CL 2025-12 conditional novelty 6.0 of 10

    A NPU-native VLM combining a MobileNet-style encoder with a hybrid Transformer-SSM backbone claims 14x lower latency and 7x lower quantization error over ViT-Transformer baselines, though quantized accuracy is not reported.

  5. Reconstruction Alignment Improves Unified Multimodal Models

    cs.CV 2025-09 conditional novelty 6.0 of 10

    RECA, a self-supervised post-training objective that conditions unified multimodal models on their own visual understanding embeddings to reconstruct input images, improves text-to-image and editing benchmarks across ...

  6. Instella-T2I: Pushing the Limits of 1D Discrete Latent Space Image Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    1D binary image latents reduce a 1024x1024 image to 128 discrete tokens and support text-to-image generation with diffusion and autoregressive models.

  7. UniCode$^2$: Cascaded Large-scale Codebooks for Unified Multimodal Understanding and Generation

    cs.CV 2025-06 conditional novelty 6.0 of 10

    UniCode² builds a 500K-entry codebook from clustered SigLIP embeddings and uses a cascaded frozen-plus-trainable codebook to unify multimodal understanding and generation with stable training and high token utilization.

  8. Are Unified Vision-Language Models Necessary: Generalization Across Understanding and Generation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    Unified vision-language models outperform task-specific models on understanding and generation, with benefits that grow with data and depend on alignment between vision input and output spaces.

  9. TokBench: Evaluating Your Visual Tokenizer before Visual Generation

    cs.CV 2025-05 conditional novelty 6.0 of 10

    TokBench measures text recognition accuracy and face similarity on reconstructed images and videos across 16 tokenizers, showing small text and faces are poorly preserved and often missed by traditional metrics.

  10. Argus-Unified: Towards A Compact and Economical Unified Model for Image Understanding and Generation

    cs.CV 2026-07 conditional novelty 5.0 of 10

    A compact unified model that reuses a frozen VLM encoder and hybrid continuous/discrete tokens reaches competitive image understanding and generation with 15.6M training images and about $2,000 in compute.

  11. UniRL: Self-Improving Unified Multimodal Models via Supervised and Reinforcement Learning

    cs.CV 2025-05 conditional novelty 5.0 of 10

    A self-improving post-training method that uses a model's own generated images as training data, with SFT and GRPO, improves generation and understanding and reduces task imbalance.

Pith tools