Flamingo: a Visual Language Model for Few-Shot Learning

TLDR

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 that can be rapidly adapted to novel tasks with few annotated examples. Flamingo achieves this by bridging pretrained vision-only and language-only models, processing arbitrarily interleaved visual and textual sequences, ingesting images or videos, and training on large multimodal web corpora to enable in-context few-shot learning across tasks such as VQA, captioning, and multiple-choice VQA. Flamingo attains state‑of‑the‑art performance on a wide range of image and video tasks with few‑shot learning, outperforming models fine‑tuned on thousands of times more task‑specific data.

Abstract

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.