


The Byte team proposed the Lynx model: multi-modal LLMs understanding cognitive generation list SoTA
Current Large Language Models (LLMs) such as GPT4 exhibit excellent multi-modal capabilities in following open instructions given an image. However, the performance of these models heavily depends on the choices of network structure, training data, and training strategies, but these choices have not been widely discussed in the previous literature. In addition, there is currently a lack of suitable benchmarks to evaluate and compare these models, which limits the development of multimodal LLMs.
Picture
- Paper: https://arxiv.org/abs/2307.02469
- Website: https://lynx-llm.github.io/
- Code: https://github.com/bytedance/lynx-llm
In this article, the author conducts a systematic and comprehensive study on the training of such models from both quantitative and qualitative aspects. More than 20 variants were set up. For the network structure, different LLMs backbones and model designs were compared; for the training data, the impact of data and sampling strategies was studied; in terms of instructions, the effect of diverse prompts on the model's instruction following ability was explored. Influence. For benchmarks, the article first proposes an open visual question answering evaluation set Open-VQA including image and video tasks.
Based on the experimental conclusions, the author proposed Lynx, which shows the most accurate multi-modal understanding compared with the existing open source GPT4-style model capabilities while maintaining the best multi-modal generation capabilities.
Evaluation scheme
Unlike typical visual language tasks, the main challenge in evaluating GPT4-style models lies in balance Performance in two aspects: text generation ability and multimodal understanding accuracy . To solve this problem, the authors propose a new benchmark Open-VQA including video and image data, and conduct a comprehensive evaluation of current open source models.
Specifically, two quantitative evaluation schemes are adopted:
- Collect open visual question answering (Open-VQA) tests Set, which contains different categories of questions on objects, OCR, counting, reasoning, action recognition, time sequencing, etc. Unlike the VQA data set, which has standard answers, Open-VQA's answers are open-ended. To evaluate the performance on Open-VQA, GPT4 is used as the discriminator, and the results are 95% consistent with human evaluation.
- In addition, the author used the OwlEval data set provided by mPLUG-owl [1] to evaluate the text generation ability of the model. Although it only contains 50 pictures and 82 questions, it covers stories Generation, ad generation, code generation and other various problems, and recruit human annotators to score the performance of different models.
Conclusion
In order to deeply study the training strategy of multi-modal LLMs, the author mainly starts from the network structure (prefix fine-tuning/cross-attention force), training data (data selection and combination ratio), instructions (single instruction/diversified instructions), LLMs model (LLaMA [5]/Vicuna [6]), image pixels (420/224) and other aspects are set With more than twenty variations, the following main conclusions have been drawn through experiments:
- #Multimodal LLMs are less capable of following instructions than LLMs. For example, InstructBLIP [2] tends to generate short replies regardless of input instructions, while other models tend to generate long sentences regardless of instructions, which the authors believe is due to a lack of high-level responses. Resulting from quality and diverse multimodal instruction data.
- #The quality of training data is crucial to the performance of the model. Based on the results of experiments on different data, it was found that using a small amount of high-quality data performs better than using large-scale noisy data. The author believes that this is the difference between generative training and contrastive training, because generative training directly learns the conditional distribution of words rather than the similarity between text and images. Therefore, for better model performance, two things need to be met in terms of data: 1) contain high-quality smooth text; 2) text and image content are well aligned.
- Quests and prompts are critical to zero-shot capabilities. Using diverse tasks and instructions can improve the model's zero-shot generation ability on unknown tasks, which is consistent with observations in plain text models.
- It is important to balance correctness with language-generating ability. If the model is undertrained on downstream tasks (such as VQA), it is more likely to generate fabricated content that does not match the visual input; while if the model is overtrained on downstream tasks, it is more likely to generate fabricated content that does not match the visual input. Short answers will not be able to generate longer answers as directed by the user.
- Prefix-finetuning (PT) is currently the best solution for multi-modal adaptation of LLMs. In experiments, the model with prefix-finetuning structure can improve the ability to follow diverse instructions faster and is easier to train than the cross-attention (CA) model structure. (prefix-tuning and cross-attention are two model structures, see the Lynx model introduction section for details)
Lynx model
The author proposed Lynx(lynx) - a prefix-finetuning GPT4-style model with two-stage training. In the first stage, approximately 120M image-text pairs are used to align visual and language embeddings; in the second stage, 20 images or videos are used for multi-modal tasks and natural language processing (NLP) ) data to adjust the model's instruction-following capabilities.
Picture
The overall structure of the Lynx model is shown in Figure 1 above.
The visual input is processed by the visual encoder to obtain visual tokens (tokens) $$W_v$$. After mapping, it is spliced with the instruction tokens $$W_l$$ as the input of LLMs. This structure is called "prefix-finetuning" in this article to distinguish it from the cross-attention structure used by Flamingo [3].
In addition, the author found that the training cost can be further reduced by adding Adapter (Adapter) after certain layers of frozen LLMs.
Model effect
The author evaluated the existing open source multi-modal LLMs model in Open-VQA, Mme [4] And the performance on OwlEval manual evaluation (results are shown in the chart below, and evaluation details are in the paper). It can be seen that the Lynx model has achieved the best performance in Open-VQA image and video understanding tasks, OwlEval manual evaluation and Mme Perception tasks. Among them, InstructBLIP also achieves high performance in most tasks, but its reply is too short. In comparison, in most cases, the Lynx model provides concise reasons to support the correct answer. Reply, which makes it more user-friendly (see the Cases display section below for some cases).
1. The indicator results on the Open-VQA image test set are shown in Table 1 below:
Picture
2. The indicator results on the Open-VQA video test set are shown in Table 2 below.
picture
3. Select the model with the top score in Open-VQA to conduct manual effect evaluation on the OwlEval evaluation set. The results are shown in Figure 4 above. It can be seen from the manual evaluation results that the Lynx model has the best language generation performance.
Picture
4. In the Mme benchmark test, the Perception class task achieved the best performance , among which 7 of 14 types of subtasks have the best performance. (See the appendix of the paper for detailed results)
Cases display
Open-VQA picture cases
OwlEval cases
Open-VQA video case
Summary
In this article, the author determined prefix-finetuning as the Open-VQA evaluation plan for the main structure of the Lynx model and open-ended answers. Experimental results show that the Lynx model performs the most accurate multi-modal understanding accuracy while maintaining the best multi-modal generation capabilities.
The above is the detailed content of The Byte team proposed the Lynx model: multi-modal LLMs understanding cognitive generation list SoTA. For more information, please follow other related articles on the PHP Chinese website!

Scientists have extensively studied human and simpler neural networks (like those in C. elegans) to understand their functionality. However, a crucial question arises: how do we adapt our own neural networks to work effectively alongside novel AI s

Google's Gemini Advanced: New Subscription Tiers on the Horizon Currently, accessing Gemini Advanced requires a $19.99/month Google One AI Premium plan. However, an Android Authority report hints at upcoming changes. Code within the latest Google P

Despite the hype surrounding advanced AI capabilities, a significant challenge lurks within enterprise AI deployments: data processing bottlenecks. While CEOs celebrate AI advancements, engineers grapple with slow query times, overloaded pipelines, a

Handling documents is no longer just about opening files in your AI projects, it’s about transforming chaos into clarity. Docs such as PDFs, PowerPoints, and Word flood our workflows in every shape and size. Retrieving structured

Harness the power of Google's Agent Development Kit (ADK) to create intelligent agents with real-world capabilities! This tutorial guides you through building conversational agents using ADK, supporting various language models like Gemini and GPT. W

summary: Small Language Model (SLM) is designed for efficiency. They are better than the Large Language Model (LLM) in resource-deficient, real-time and privacy-sensitive environments. Best for focus-based tasks, especially where domain specificity, controllability, and interpretability are more important than general knowledge or creativity. SLMs are not a replacement for LLMs, but they are ideal when precision, speed and cost-effectiveness are critical. Technology helps us achieve more with fewer resources. It has always been a promoter, not a driver. From the steam engine era to the Internet bubble era, the power of technology lies in the extent to which it helps us solve problems. Artificial intelligence (AI) and more recently generative AI are no exception

Harness the Power of Google Gemini for Computer Vision: A Comprehensive Guide Google Gemini, a leading AI chatbot, extends its capabilities beyond conversation to encompass powerful computer vision functionalities. This guide details how to utilize

The AI landscape of 2025 is electrifying with the arrival of Google's Gemini 2.0 Flash and OpenAI's o4-mini. These cutting-edge models, launched weeks apart, boast comparable advanced features and impressive benchmark scores. This in-depth compariso


Hot AI Tools

Undresser.AI Undress
AI-powered app for creating realistic nude photos

AI Clothes Remover
Online AI tool for removing clothes from photos.

Undress AI Tool
Undress images for free

Clothoff.io
AI clothes remover

Video Face Swap
Swap faces in any video effortlessly with our completely free AI face swap tool!

Hot Article

Hot Tools

SublimeText3 English version
Recommended: Win version, supports code prompts!

SecLists
SecLists is the ultimate security tester's companion. It is a collection of various types of lists that are frequently used during security assessments, all in one place. SecLists helps make security testing more efficient and productive by conveniently providing all the lists a security tester might need. List types include usernames, passwords, URLs, fuzzing payloads, sensitive data patterns, web shells, and more. The tester can simply pull this repository onto a new test machine and he will have access to every type of list he needs.

Dreamweaver Mac version
Visual web development tools

Notepad++7.3.1
Easy-to-use and free code editor

PhpStorm Mac version
The latest (2018.2.1) professional PHP integrated development tool
