Flamingo: a Visual Language Model for Few-Shot Learning
Flamingo: визуальная языковая модель для обучения по малому числу примеров
2022-04-29
SCID: 54.1/axxm8y2q
Discuss with AI
Few-shot learningIn-context learningMultimodal web corporaVisual language modelsVisual question answering
Figures from the paper
Abstract (AI)
Building models that can be rapidly adapted to novel tasks using only a handful of annotated examples is an open challenge for multimodal machine learning research. We introduce Flamingo, a family of Visual Language Models (VLM) with this ability. We propose key architectural innovations to: (i) bridge powerful pretrained vision-only and language-only models, (ii) handle sequences of arbitrarily interleaved visual and textual data, and (iii) seamlessly ingest images or videos as inputs. Thanks to their flexibility, Flamingo models can be trained on large-scale multimodal web corpora containing arbitrarily interleaved text and images, which is key to endow them with in-context few-shot learning capabilities. We perform a thorough evaluation of our models, exploring and measuring their ability to rapidly adapt to a variety of image and video tasks. These include open-ended tasks such as visual question-answering, where the model is prompted with a question which it has to answer; captioning tasks, which evaluate the ability to describe a scene or an event; and close-ended tasks such as multiple-choice visual question-answering. For tasks lying anywhere on this spectrum, a single Flamingo model can achieve a new state of the art with few-shot learning, simply by prompting the model with task-specific examples. On numerous benchmarks, Flamingo outperforms models fine-tuned on thousands of times more task-specific data.
Key Findings
1
A single Flamingo model achieves state-of-the-art few-shot performance on open-ended, captioning, and multiple-choice visual question-answering tasks through task-specific prompting.
2
Flamingo is a family of visual language models designed to adapt rapidly to novel multimodal tasks using only a few annotated examples.
3
Flamingo outperforms models fine-tuned on thousands of times more task-specific data across numerous benchmarks.
4
The architecture bridges pretrained vision-only and language-only models and processes arbitrarily interleaved visual and textual sequences, including images and videos.
5
Training on large-scale multimodal web corpora enables Flamingo to perform in-context few-shot learning across diverse vision-language tasks.
Research Object
Flamingo family of Visual Language Models (VLMs)
Research Subject
In-context few-shot adaptation and performance across image- and video-based vision-language tasks
Publication Details
Publication Date
2022-04-29
Journal
Publisher
ISSN
Cited by
1555
Open access PDF
Access Type
Author Information
Download PDF
Subscribe to digest
Cited by15
A Comprehensive Overview of Large Language Models2025
Multimodal Learning With Transformers: A Survey2023
Talking about Large Language Models2024
A survey on LLM-based multi-agent systems: workflow, infrastructure, and challenges2024
A Comprehensive Review of Deep Learning: Architectures, Recent Advances, and Applications2024
Vision-language models for medical report generation and visual question answering: a review2024
Review of large vision models and visual prompt engineering2023
Driving with LLMs: Fusing Object-Level Vector Modality for Explainable Autonomous Driving2024
Interactive computer-aided diagnosis on medical image using large language models2024
Alpha-CLIP: A CLIP Model Focusing on Wherever you Want2024
When Large Language Models Meet Vector Databases: A Survey2025
Large Language Models: A Structured Taxonomy and Review of Challenges, Limitations, Solutions, and Future Directions2025
Benchmark Evaluations, Applications, and Challenges of Large Vision Language Models: A Survey2025
(Mis)Matching Metadata: Improving Accessibility in Digital Visual Archives through the EyCon Project2023
Large language models for clinical artificial intelligence in healthcare a systematic review2026