# TIME: Text and Image Mutual-Translation Adversarial Networks

arXiv: https://arxiv.org/abs/2005.13192 · HTML: https://mlsanity.com/paper/2005.13192/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 27 May 2020.

## Built on by (most cited first, 9 of 9)

- [Towards Faster and Stabilized GAN Training for High-fidelity Few-shot Image Synthesis](https://mlsanity.com/paper/2101.04775/) (2021)
- [GALIP: Generative Adversarial CLIPs for Text-to-Image Synthesis](https://mlsanity.com/paper/2301.12959/) (2023)
- [Attribute-Centric Compositional Text-to-Image Generation](https://mlsanity.com/paper/2301.01413/) (2023)
- [DSE-GAN: Dynamic Semantic Evolution Generative Adversarial Network for Text-to-Image Generation](https://mlsanity.com/paper/2209.01339/) (2022)
- [Layout-Bridging Text-to-Image Synthesis](https://mlsanity.com/paper/2208.06162/) (2022)
- [Fine-grained Cross-modal Fusion based Refinement for Text-to-Image Synthesis](https://mlsanity.com/paper/2302.08706/) (2023)
- [End-to-end Training for Text-to-Image Synthesis using Dual-Text Embeddings](https://mlsanity.com/paper/2502.01507/) (2025)
- [A Simple and Effective Baseline for Attentional Generative Adversarial Networks](https://mlsanity.com/paper/2306.14708/) (2023)
- [JEPA-T: Joint-Embedding Predictive Architecture with Text Fusion for Image Generation](https://mlsanity.com/paper/2510.00974/) (2025)

## Abstract (arXiv)

Focusing on text-to-image (T2I) generation, we propose Text and Image Mutual-Translation Adversarial Networks (TIME), a lightweight but effective model that jointly learns a T2I generator G and an image captioning discriminator D under the Generative Adversarial Network framework. While previous methods tackle the T2I problem as a uni-directional task and use pre-trained language models to enforce the image--text consistency, TIME requires neither extra modules nor pre-training. We show that the performance of G can be boosted substantially by training it jointly with D as a language model. Specifically, we adopt Transformers to model the cross-modal connections between the image features and word embeddings, and design an annealing conditional hinge loss that dynamically balances the adversarial learning. In our experiments, TIME achieves state-of-the-art (SOTA) performance on the CUB and MS-COCO dataset (Inception Score of 4.91 and Fr'echet Inception Distance of 14.3 on CUB), and shows promising performance on MS-COCO on image captioning and downstream vision-language tasks.