Research map / Image generation / GANs
Adversarial Feature Matching for Text Generation
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-space MMD matching | TextGAN · BookCorpus + arXiv test | BLEU-4: 0.04$\pm$0.04 -> 0.13$\pm$0.05 | Helped |
| Replace feature-space MMD with mean matching (ablation) | TextGAN · BookCorpus + arXiv test | BLEU-4: 0.13$\pm$0.05 -> 0.09$\pm$0.04 | Hurt |
| Replace feature-space MMD with covariance matching (ablation) | TextGAN · BookCorpus + arXiv test | BLEU-4: 0.13$\pm$0.05 -> 0.12$\pm$0.03 | No clear effect |
| Compress the MMD feature space to 200 dimensions (ablation) | TextGAN · BookCorpus + arXiv test | BLEU-4: 0.13$\pm$0.05 -> 0.11$\pm$0.05 | No clear effect |
| Compress the MMD feature space to 200 dimensions (ablation) | TextGAN · BookCorpus + arXiv test | BLEU-3: 0.49$\pm$0.06 -> 0.52$\pm$0.07 | Helped |
| Compress the MMD feature space to 200 dimensions (ablation) | TextGAN · BookCorpus + arXiv test | BLEU-4: 0.13$\pm$0.05 -> 0.11$\pm$0.05 | Hurt |
Closest papers by meaning
- Adversarial Text Generation via Feature-Mover's Distance (NeurIPS 2018 · similarity 0.92)
- Long Text Generation via Adversarial Training with Leaked Information (AAAI 2018 · similarity 0.91)
- Self-Adversarial Learning with Comparative Discrimination for Text Generation (ICLR 2020 · similarity 0.90)
- TextGAIL: Generative Adversarial Imitation Learning for Text Generation (AAAI 2021 · similarity 0.90)
- Branch-GAN: Improving Text Generation with (not so) Large Language Models (ICLR 2024 · similarity 0.89)
- Adversarial Text to Continuous Image Generation (CVPR 2024 · similarity 0.86)
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.