Back To Projects

Generating Instagram Captions with ViT-GPT2 and GPT3

Ariel M.

This paper presents a novel approach for generating Instagram captions based on visual features and language models. Our caption generator combines Vision Transformer-GPT2 and GPT3 to generate descriptive and engaging captions in the style of an Instagram post.


This paper presents a novel approach for generating Instagram captions based on visual features and language models. Our caption generator combines Vision Transformer-GPT2 and GPT3 to generate descriptive and engaging captions in the style of an Instagram post. Our caption generator has the potential to assist social media users, content creators, and marketers in generating high-quality and engaging captions for their Instagram posts. Future work includes exploring more advanced visual and textual features and incorporating user feedback and preferences. Instagram is a popular social media platform for sharing photos and videos. One of the key elements of a successful Instagram post is a catchy caption that enhances the visual content and engages the audience. However, coming up with a good caption can be challenging, especially for users who are not native speakers of the language or lack creativity. In this paper, we propose an automatic caption generator that takes an image as input and generates an Instagram-style caption using state-of-the-art language models.

Ariel M.
Roger Jin
Computer Science MS at MIT, ML researcher at Apple

Related Projects

Is That Slang?

How can we better understand Gen-Z slang that is being used today? In order to tackle this challenge, I decided to make a NLP model that processes sentences and determines whether or not they contain Gen-Z slang.
Dhaniya V.
Mentored by Matan Gans
Fake News Detection with BERT

In this paper, we propose a fake news detection model using a fine-tuned BERT (Bidirectional Encoder Representations from Transformers) model.
David S.
Mentored by Roger Jin
From Text to Visuals: Leveraging AI for Immersive Storytelling

This research investigates the use of AI to generate visual representations of Edgar Allan Poe's literary works, aiming to enhance reader engagement by bridging traditional literature with modern digital media through AI-generated imagery, while identifying challenges related to visual consistency and narrative complexity.
Ian M.
Mentored by Alaisha Alexander