SPAE: Semantic Pyramid AutoEncoder for Multimodal Generation with Frozen
LLMs
- URL: http://arxiv.org/abs/2306.17842v3
- Date: Sat, 28 Oct 2023 18:09:46 GMT
- Title: SPAE: Semantic Pyramid AutoEncoder for Multimodal Generation with Frozen
LLMs
- Authors: Lijun Yu, Yong Cheng, Zhiruo Wang, Vivek Kumar, Wolfgang Macherey,
Yanping Huang, David A. Ross, Irfan Essa, Yonatan Bisk, Ming-Hsuan Yang,
Kevin Murphy, Alexander G. Hauptmann, Lu Jiang
- Abstract summary: We introduce SPAE for enabling frozen LLMs to perform both understanding and generation tasks involving non-linguistic modalities such as images or videos.
The resulting lexical tokens capture both the semantic meaning and the fine-grained details needed for visual reconstruction.
Our method marks the first successful attempt to enable a frozen LLM to generate image content while surpassing state-of-the-art performance in image understanding tasks, under the same setting, by over 25%.
- Score: 124.29233620842462
- License: http://creativecommons.org/licenses/by/4.0/
- Abstract: In this work, we introduce Semantic Pyramid AutoEncoder (SPAE) for enabling
frozen LLMs to perform both understanding and generation tasks involving
non-linguistic modalities such as images or videos. SPAE converts between raw
pixels and interpretable lexical tokens (or words) extracted from the LLM's
vocabulary. The resulting tokens capture both the semantic meaning and the
fine-grained details needed for visual reconstruction, effectively translating
the visual content into a language comprehensible to the LLM, and empowering it
to perform a wide array of multimodal tasks. Our approach is validated through
in-context learning experiments with frozen PaLM 2 and GPT 3.5 on a diverse set
of image understanding and generation tasks. Our method marks the first
successful attempt to enable a frozen LLM to generate image content while
surpassing state-of-the-art performance in image understanding tasks, under the
same setting, by over 25%.
Related papers
- Paying More Attention to Image: A Training-Free Method for Alleviating Hallucination in LVLMs [14.381188702947949]
Large Vision-Language Models (LVLMs) primarily align image features of vision encoder with Large Language Models (LLMs) to leverage their superior text generation capabilities.
This imbalance in LVLMs may result in the instances of hallucinatory.
We introduce a training-free algorithm to find an equilibrium point between image comprehension and language inference.
arXiv Detail & Related papers (2024-07-31T17:46:57Z) - Auto-Encoding Morph-Tokens for Multimodal LLM [151.2618346912529]
We propose encoding images into morph-tokens to serve a dual purpose: for comprehension, they act as visual prompts instructing MLLM to generate texts.
Experiments show that morph-tokens can achieve a new SOTA for multimodal comprehension and generation simultaneously.
arXiv Detail & Related papers (2024-05-03T08:43:06Z) - Beyond Text: Frozen Large Language Models in Visual Signal Comprehension [34.398976855955404]
Vision-to-Language Tokenizer, abbreviated as V2T Tokenizer, transforms an image into a foreign language'' with the combined aid of an encoder-decoder, the LLM vocabulary, and a CLIP model.
We undertake rigorous experiments to validate our method, encompassing understanding tasks like image recognition, image captioning, and visual question answering, as well as image denoising tasks like inpainting, outpainting, deblurring, and shift restoration.
arXiv Detail & Related papers (2024-03-12T17:59:51Z) - Browse and Concentrate: Comprehending Multimodal Content via prior-LLM Context Fusion [70.9767518332692]
Multimodal Large Language Models (MLLMs) that incorporate LLMs with pre-trained vision models have recently demonstrated impressive performance across diverse vision-language tasks.
However, they fall short to comprehend context involving multiple images.
We propose a two phase paradigm, browse-and-concentrate, to enable in-depth multimodal context fusion.
arXiv Detail & Related papers (2024-02-19T14:59:07Z) - MLLMs-Augmented Visual-Language Representation Learning [70.5293060238008]
We demonstrate that Multi-modal Large Language Models (MLLMs) can enhance visual-language representation learning.
Our approach is simple, utilizing MLLMs to extend multiple diverse captions for each image.
We propose "text shearing" to maintain the quality and availability of extended captions.
arXiv Detail & Related papers (2023-11-30T18:05:52Z) - Frozen Transformers in Language Models Are Effective Visual Encoder Layers [26.759544759745648]
Large language models (LLMs) are surprisingly strong encoders for purely visual tasks in the absence of language.
Our work pushes the boundaries of leveraging LLMs for computer vision tasks.
We propose the information filtering hypothesis to explain the effectiveness of pre-trained LLMs in visual encoding.
arXiv Detail & Related papers (2023-10-19T17:59:05Z) - BuboGPT: Enabling Visual Grounding in Multi-Modal LLMs [101.50522135049198]
BuboGPT is a multi-modal LLM with visual grounding that can perform cross-modal interaction between vision, audio and language.
Our contributions are two-fold: 1) An off-the-shelf visual grounding module based on SAM that extracts entities in a sentence and find corresponding masks in the image.
Our experiments show that BuboGPT achieves impressive multi-modality understanding and visual grounding abilities during the interaction with human.
arXiv Detail & Related papers (2023-07-17T15:51:47Z) - LLM-CXR: Instruction-Finetuned LLM for CXR Image Understanding and Generation [51.08810811457617]
vision-language alignment in LLMs is actively being researched to enable multimodal reasoning and visual IO.
We develop a method for instruction-tuning an LLM only on text to gain vision-language capabilities for medical images.
Our model, LLM-CXR, trained in this approach shows better image-text alignment in both CXR understanding and generation tasks.
arXiv Detail & Related papers (2023-05-19T07:44:39Z)
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.