Research map / Image generation / GANs
Adversarial Text Generation via Feature-Mover's Distance
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| use feature-mover's distance critic (FM-GAN) instead of RL-based or other GAN text generators | FM-GAN · EMNLP WMT News | human score: 3.41 -> 3.72 vs LeakGAN | Helped |
| replace cross-aligned autoencoder GAN loss with FMD critic | cross-aligned autoencoder style transfer · Yelp sentiment transfer | sentiment accuracy (%): 80.6 -> 89.8 | Helped |
| FM-GAN versus Controllable model with pretrained sentiment classifier | style transfer · Yelp sentiment transfer | content preservation rating: 4.6 -> 4.5 | Hurt |
| replace CipherGAN critic with FMD objective | CipherGAN · Brown-W200 Vigenere decipher | deciphering accuracy (%): 75.7 -> 77.2 | Helped |
Closest papers by meaning
- Adversarial Feature Matching for Text Generation (ICML 2017 · similarity 0.92)
- Branch-GAN: Improving Text Generation with (not so) Large Language Models (ICLR 2024 · similarity 0.90)
- Long Text Generation via Adversarial Training with Leaked Information (AAAI 2018 · similarity 0.89)
- Self-Adversarial Learning with Comparative Discrimination for Text Generation (ICLR 2020 · similarity 0.89)
- TextGAIL: Generative Adversarial Imitation Learning for Text Generation (AAAI 2021 · similarity 0.88)
- Adversarial Text to Continuous Image Generation (CVPR 2024 · similarity 0.85)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.