Research map / Probabilistic and scientific modeling / Variational and generative sampling
DialogWAE: Multimodal Response Generation with Conditional Wasserstein Auto-Encoder
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 a Gaussian mixture prior instead of a single Gaussian prior (ablation) | DialogWAE · SwitchBoard | BLEU recall: 0.394 -> 0.420 | Helped |
| Use a Gaussian mixture prior instead of a single Gaussian prior (ablation) | DialogWAE · DailyDialog | BLEU recall: 0.341 -> 0.372 | Helped |
| Use a Gaussian mixture prior instead of a single Gaussian prior (ablation) | DialogWAE · SwitchBoard | inter-dist dist-2: 0.413 -> 0.555 | Helped |
| Use a Gaussian mixture prior instead of a single Gaussian prior (ablation) | DialogWAE · DailyDialog | inter-dist dist-2: 0.583 -> 0.597 | Helped |
| Vary the number of Gaussian prior components from 1 to 9 (ablation) | DialogWAE-GMP · SwitchBoard and DailyDialog | performance (reported across metrics) | Mixed |
| Use around three Gaussian prior components (ablation) | DialogWAE-GMP · SwitchBoard and DailyDialog | performance (reported across metrics) | Helped |
| Use DialogWAE-GMP rather than VHCR | DailyDialog · human-rated coherence | coherence: 26.8% -> 31.6% | Helped |
Closest papers by meaning
- Towards Diverse, Relevant and Coherent Open-Domain Dialogue Generation via Hybrid Latent Variables (AAAI 2023 · similarity 0.89)
- Improving Variational Encoder-Decoders in Dialogue Generation (AAAI 2018 · similarity 0.88)
- Best of Both Worlds: Transferring Knowledge from Discriminative Learning to a Generative Visual Dialog Model (NeurIPS 2017 · similarity 0.88)
- Learning towards Selective Data Augmentation for Dialogue Generation (AAAI 2023 · similarity 0.88)
- V^2Dial: Unification of Video and Visual Dialog via Multimodal Experts (CVPR 2025 · similarity 0.86)
- Structure-Aware Multimodal Sequential Learning for Visual Dialog (AAAI 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.