search
HomeTechnology peripheralsAIBytedance Doubao and Wuhan University proposed CAL: enhancing multi-modal alignment effects through visually related tokens

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果
The AIxiv column is a column where this site publishes academic and technical content. In the past few years, the AIxiv column of this site has received more than 2,000 reports, covering top laboratories from major universities and companies around the world, effectively promoting academic exchanges and dissemination. If you have excellent work that you want to share, please feel free to contribute or contact us for reporting. Submission email: liyazhou@jiqizhixin.com; zhaoyunfeng@jiqizhixin.com

The current mainstream visual language model (VLM) is mainly based on the large language model (LLM) for further fine-tuning. Therefore, it is necessary to map the image to the embedding space of LLM in various ways, and then use autoregressive methods to predict the answer based on the image token.

In this process, modal alignment is implicitly implemented through text tokens. How to align this step well is very critical.

In response to this problem, researchers from Wuhan University, ByteDance Beanbao Large Model Team and the University of Chinese Academy of Sciences proposed a text token screening method (CAL) based on contrastive learning to screen out text Tokens that are highly related to the image are increased in weight of the loss function to achieve more accurate multi-modal alignment.

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果

  • Paper link: https://arxiv.org/pdf/2405.17871
  • Code link: https://github.com/foundation-multimodal-models/CAL

CAL has the following highlights:

  • can be directly nested into the training process without additional pre-training stage.
  • has achieved significant improvements in OCR and Caption benchmarks. From the visualization, it can be found that CAL makes the image modal alignment better.
  • CAL makes the training process more resistant to noisy data.

Research motivation

Currently, visual language models rely on the alignment of image modalities, and how to do alignment is very critical. The current mainstream method is to perform implicit alignment through text autoregression, but the contribution of each text token to image alignment is inconsistent. It is very necessary to distinguish these text tokens.

CAL proposed that in the existing visual language model (VLM) training data, text tokens can be divided into three categories:

  • Text that is highly related to pictures: such as entities ( Such as people, animals, objects), quantity, color, text, etc. These tokens directly correspond to image information and are crucial for multi-modal alignment.
  • Text with low correlation to the picture: Such as following words or content that can be inferred from the previous text. These tokens are actually mainly used to train the plain text capabilities of VLM.
  • Text that contradicts the image content: These tokens are inconsistent with image information and may even provide misleading information, negatively affecting the multi-modal alignment process.

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果标 Figure 1: The green mark is related to the high -related Token, the red is the contrary to the content, and the colorless is the neutral Token

During the training process, the latter two types of token are actually actually occupy a larger proportion, but because they are not strongly dependent on the image, they have little effect on the modal alignment of the image. Therefore, in order to achieve better alignment, it is necessary to increase the weight of the first type of text tokens, that is, the tokens that are highly related to the image. How to find this part of the token has become the key to solving this problem.

Method

Finding tokens that are highly related to the image This problem can be solved by condition contrastive.
For each image-text pair in the training data, without image input, the logit on each text token represents LLM’s estimate of the occurrence of this situation based on context and existing knowledge. value.
  • If you add image input in front, it is equivalent to providing additional contextual information. In this case, the logit of each text token will be adjusted based on the new situation. The logit changes in these two cases represent the impact of the new condition of the picture on each text token.
  • Specifically, during the training process, CAL inputs the image and text sequences and individual text sequences into the large language model (LLM) respectively to obtain the logit of each text token. By calculating the logit difference between these two cases, we can measure the impact of the image on each token. The larger the logit difference, the greater the impact of the image on the token, so the token is more relevant to the image. The figure below shows the flow chart of the logit diff and CAL methods for text tokens.对 Figure 2: The left picture is the visualization of the token logit diff in the two situations. The picture on the right is the visualization of the CAL method process

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果Cal in Llava Experimental verification was conducted on two mainstream models: MGM and MGM, and performance improvements were achieved in models of different sizes.
Contains the following four parts of verification:

(1) Models using CAL perform better on various benchmark indicators.


(2) Create a batch of noise data (image-text mismatch) by randomly exchanging the text in the two image-text pairs in proportion and use it for model training. CAL makes the training process Has stronger data anti-noise performance.度 Figure 3: In the case of noise training at different intensity, the performance of CAL and the baseline

(3) calculates the attention scores of the picture token in the answer part of QA Case, And plotting it on the original image, the CAL-trained model has a clearer attention distribution map.

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果

C Figure 4: The baseline and CAL's Attention Map can be visualized. The right side of each pair is CAL
(4) to the text token to the text token in its most similar LLM vocabulary. Drawing it onto the original image, the CAL-trained model mapping content is closer to the image content. A Figure 5: Imam into the Image Token as the most similar vocabulary, and correspond to the original picture. The model team was established in 2023 and is committed to developing the industry's most advanced AI large model technology, becoming a world-class research team, and contributing to technological and social development.

字节豆包、武大提出 CAL:通过视觉相关的 token 增强多模态对齐效果
The Doubao Big Model team has long-term vision and determination in the field of AI. Its research directions cover NLP, CV, speech, etc., and it has laboratories and research positions in China, Singapore, the United States and other places. Relying on the platform's sufficient data, computing and other resources, the team continues to invest in related fields. It has launched a self-developed general large model to provide multi-modal capabilities. The downstream supports 50+ businesses such as Doubao, Buttons, and Jimeng, and is open to the volcano engine. Corporate customers. At present, Doubao APP has become the AIGC application with the largest number of users in the Chinese market. Welcome to join the ByteDance Beanbao model team.

The above is the detailed content of Bytedance Doubao and Wuhan University proposed CAL: enhancing multi-modal alignment effects through visually related tokens. For more information, please follow other related articles on the PHP Chinese website!

Statement
The content of this article is voluntarily contributed by netizens, and the copyright belongs to the original author. This site does not assume corresponding legal responsibility. If you find any content suspected of plagiarism or infringement, please contact admin@php.cn
字节跳动旗下视频编辑 App CapCut 全球用户总支出超 1 亿美元字节跳动旗下视频编辑 App CapCut 全球用户总支出超 1 亿美元Sep 14, 2023 pm 09:41 PM

字节跳动旗下的创意视频剪辑工具CapCut在中国、美国和东南亚拥有大量用户。该工具支持安卓、iOS和PC平台市场调研机构data.ai最新报告指出,截至2023年9月11日,CapCut在iOS和GooglePlay上的用户总支出已突破1亿美元(本站备注:当前约7.28亿元人民币),成功超越Splice(2022年下半年排名第一)成为2023年上半年全球最吸金的视频剪辑应用,与2022年下半年相比增长了180%。截至2023年8月,全球有4.9亿人通过iPhone和安卓手机使用CapCut。da

深圳字节跳动后海中心总建筑面积 7.74 万平方米完成主体结构封顶深圳字节跳动后海中心总建筑面积 7.74 万平方米完成主体结构封顶Jan 24, 2024 pm 05:27 PM

据南山区政府官方微信公众号“创新南山”透露,深圳字节跳动后海中心项目最近取得了重要进展。根据中建一局建设发展公司的消息,该项目主体结构提前3天全部完成封顶工作。这一消息意味着南山后海核心区将迎来一个新的地标建筑。深圳字节跳动后海中心项目位于南山区后海核心区,是今日头条科技有限公司在深圳市的总部办公大楼。总建筑面积为7.74万平方米,高约150米,共有地下4层和地上32层。据悉,深圳字节跳动后海中心项目将成为一座创新型超高层建筑,集办公、娱乐、餐饮等功能为一体。该项目将有助于深圳推动互联网产业的集

字节跳动模型大规模部署实战字节跳动模型大规模部署实战Apr 12, 2023 pm 08:31 PM

一. 背景介绍在字节跳动,基于深度学习的应用遍地开花,工程师关注模型效果的同时也需要关注线上服务一致性和性能,早期这通常需要算法专家和工程专家分工合作并紧密配合来完成,这种模式存在比较高的 diff 排查验证等成本。随着 PyTorch/TensorFlow 框架的流行,深度学习模型训练和在线推理完成了统一,开发者仅需要关注具体算法逻辑,调用框架的 Python API 完成训练验证过程即可,之后模型可以很方便的序列化导出,并由统一的高性能 C++ 引擎完成推理工作。提升了开发者训练到部署的体验

NUS和字节跨界合作,通过模型优化实现训练提速72倍,并荣获AAAI2023杰出论文。NUS和字节跨界合作,通过模型优化实现训练提速72倍,并荣获AAAI2023杰出论文。May 06, 2023 pm 10:46 PM

近日,人工智能国际顶会AAAI2023公布评选结果。新加坡国立大学(NUS)与字节跳动机器学习团队(AML)合作的CowClip技术论文入围杰出论文(DistinguishedPapers)。CowClip是一项模型训练优化策略,可以在保证模型精度的前提下,实现在单张GPU上的模型训练速度提升72倍,相关代码现已开源。​论文地址:https://arxiv.org/abs/2204.06240​开源地址:https://github.com/bytedance/LargeBatchCTR​AAA

字节跳动拓展全球研发中心,派遣工程师加拿大和澳大利亚等地字节跳动拓展全球研发中心,派遣工程师加拿大和澳大利亚等地Jan 18, 2024 pm 04:00 PM

IT之家1月18日消息,针对近日TikTok国内员工转岗海外的传言,据接近字节跳动的人士透露,该公司正在加拿大、澳大利亚等地筹建研发中心。目前,部分研发中心已试运营半年左右,未来将支持TikTok、CapCut、Lemon8等多个海外业务研发。字节跳动计划以当地招聘为主,并辅助少量外派的方式筹建相关研发中心。据了解,过去半年,该公司已从美国、中国、新加坡等地选派少量工程师参与筹建。其中,从中国向两地研发中心累计派出包括产品、研发和运营岗位120人。相关人士表示,此举是为了应对海外业务的发展,更好

PICO 4 销量远远低于预期,消息称字节跳动将取消下一代 VR 头显 PICO 5PICO 4 销量远远低于预期,消息称字节跳动将取消下一代 VR 头显 PICO 5Dec 15, 2023 am 09:34 AM

本站12月13日消息,据TheInformation,字节跳动准备砍掉其PICO新一代VR头显PICO5,因为现款PICO4的销量远远低于预期。根据EqualOcean在今年10月的一篇文章,据称字节跳动将逐步关闭PICO,并放弃元宇宙领域。文章指出,字节跳动认为PICO所处的硬件领域并非其专长,几年来的成绩未达到预期,并且对未来缺乏希望在当时,字节跳动的相关负责人对于关于“逐步放弃PICO业务”的传闻进行了回应,称这一消息是不实的。他们表示PICO业务仍在正常运营,并且公司将会长期投入扩展现实

Pico疑似即将发布全新VR头显Pico 4S,硬件升级引期待Pico疑似即将发布全新VR头显Pico 4S,硬件升级引期待Mar 16, 2024 pm 08:49 PM

近期,科技圈再次掀起了一股虚拟现实(VR)的热潮。据称,字节跳动旗下的VR子公司Pico即将推出全新的独立VR头显——Pico4S。一位名为@Lunayian的用户在社交媒体上发布了一张3D模型图片,声称该图片来自PicoConnectPC客户端,展示了Pico4S的右控制器设计。这款控制器的外观与去年9月在网络上泄露的"Pico5"控制器非常相似,但与Pico4的控制器有一些明显的差异,主要体现在取消了定位环。这一设计调整可能预示着Pico4S将带来全新的用户体验和交互方式。据了解,Pico在

抖音子公司推出基于云雀模型的 AI 机器人“豆包”抖音子公司推出基于云雀模型的 AI 机器人“豆包”Aug 23, 2023 am 10:53 AM

本站8月17日消息,字节跳动旗下LLM人工智能机器人“豆包”现已开始小范围邀请测试,用户可通过手机号、抖音或者AppleID登录。根据报道,据称字节跳动公司开发了一款名为"豆包"的AI工具,该工具基于云雀模型,提供聊天机器人、写作助手和英语学习助手等功能。它可以回答各种问题并进行对话,帮助人们获取信息。"豆包"支持网页Web平台、iOS和安卓平台,但在iOS平台上需要通过TestFlight进行安装官网用户协议显示,“豆包”软件及相关服务系指北京春田知韵科

See all articles

Hot AI Tools

Undresser.AI Undress

Undresser.AI Undress

AI-powered app for creating realistic nude photos

AI Clothes Remover

AI Clothes Remover

Online AI tool for removing clothes from photos.

Undress AI Tool

Undress AI Tool

Undress images for free

Clothoff.io

Clothoff.io

AI clothes remover

AI Hentai Generator

AI Hentai Generator

Generate AI Hentai for free.

Hot Article

Repo: How To Revive Teammates
1 months agoBy尊渡假赌尊渡假赌尊渡假赌
R.E.P.O. Energy Crystals Explained and What They Do (Yellow Crystal)
2 weeks agoBy尊渡假赌尊渡假赌尊渡假赌
Hello Kitty Island Adventure: How To Get Giant Seeds
1 months agoBy尊渡假赌尊渡假赌尊渡假赌

Hot Tools

Dreamweaver Mac version

Dreamweaver Mac version

Visual web development tools

MinGW - Minimalist GNU for Windows

MinGW - Minimalist GNU for Windows

This project is in the process of being migrated to osdn.net/projects/mingw, you can continue to follow us there. MinGW: A native Windows port of the GNU Compiler Collection (GCC), freely distributable import libraries and header files for building native Windows applications; includes extensions to the MSVC runtime to support C99 functionality. All MinGW software can run on 64-bit Windows platforms.

MantisBT

MantisBT

Mantis is an easy-to-deploy web-based defect tracking tool designed to aid in product defect tracking. It requires PHP, MySQL and a web server. Check out our demo and hosting services.

Atom editor mac version download

Atom editor mac version download

The most popular open source editor

Notepad++7.3.1

Notepad++7.3.1

Easy-to-use and free code editor