Saved in:
Bibliographic Details
Main Authors: Nabati, Ofir, Tennenholtz, Guy, Hsu, ChihWei, Ryu, Moonkyung, Ramachandran, Deepak, Chow, Yinlam, Li, Xiang, Boutilier, Craig
Format: Preprint
Published: 2024
Subjects:
Online Access:https://arxiv.org/abs/2412.10419
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908382273732608
author Nabati, Ofir
Tennenholtz, Guy
Hsu, ChihWei
Ryu, Moonkyung
Ramachandran, Deepak
Chow, Yinlam
Li, Xiang
Boutilier, Craig
author_facet Nabati, Ofir
Tennenholtz, Guy
Hsu, ChihWei
Ryu, Moonkyung
Ramachandran, Deepak
Chow, Yinlam
Li, Xiang
Boutilier, Craig
contents We address the problem of interactive text-to-image (T2I) generation, designing a reinforcement learning (RL) agent which iteratively improves a set of generated images for a user through a sequence of prompt expansions. Using human raters, we create a novel dataset of sequential preferences, which we leverage, together with large-scale open-source (non-sequential) datasets. We construct user-preference and user-choice models using an EM strategy and identify varying user preference types. We then leverage a large multimodal language model (LMM) and a value-based RL approach to suggest an adaptive and diverse slate of prompt expansions to the user. Our Preference Adaptive and Sequential Text-to-image Agent (PASTA) extends T2I models with adaptive multi-turn capabilities, fostering collaborative co-creation and addressing uncertainty or underspecification in a user's intent. We evaluate PASTA using human raters, showing significant improvement compared to baseline methods. We also open-source our sequential rater dataset and simulated user-rater interactions to support future research in user-centric multi-turn T2I systems.
format Preprint
id arxiv_https___arxiv_org_abs_2412_10419
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Preference Adaptive and Sequential Text-to-Image Generation
Nabati, Ofir
Tennenholtz, Guy
Hsu, ChihWei
Ryu, Moonkyung
Ramachandran, Deepak
Chow, Yinlam
Li, Xiang
Boutilier, Craig
Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
Systems and Control
We address the problem of interactive text-to-image (T2I) generation, designing a reinforcement learning (RL) agent which iteratively improves a set of generated images for a user through a sequence of prompt expansions. Using human raters, we create a novel dataset of sequential preferences, which we leverage, together with large-scale open-source (non-sequential) datasets. We construct user-preference and user-choice models using an EM strategy and identify varying user preference types. We then leverage a large multimodal language model (LMM) and a value-based RL approach to suggest an adaptive and diverse slate of prompt expansions to the user. Our Preference Adaptive and Sequential Text-to-image Agent (PASTA) extends T2I models with adaptive multi-turn capabilities, fostering collaborative co-creation and addressing uncertainty or underspecification in a user's intent. We evaluate PASTA using human raters, showing significant improvement compared to baseline methods. We also open-source our sequential rater dataset and simulated user-rater interactions to support future research in user-centric multi-turn T2I systems.
title Preference Adaptive and Sequential Text-to-Image Generation
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Computation and Language
Machine Learning
Systems and Control
url https://arxiv.org/abs/2412.10419