Research map / Image generation / Autoregressive and latent generation
Learning To Memorize Feature Hallucination for One-Shot Image Generation
Check 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 MFH instead of the compared one-shot generation methods | One-shot image generation · AnimalFace | FID: 102.31 -> 75.28 | Helped |
| Use MFH instead of the compared one-shot generation methods | One-shot image generation · NABirds | FID: 84.56 -> 42.24 | Helped |
| Add MFH-generated images as data augmentation | ProtoNet · NABirds 5-way 1-shot | accuracy: 77.93±0.67 -> 79.02±0.61 | Helped |
| Add MFH-generated images as data augmentation | MetaIRNet · NABirds 5-way 1-shot | accuracy: 79.21±0.63 -> 82.98±0.60 | Helped |
| Add flip augmentation | ProtoNet · NABirds 5-way 1-shot | accuracy: 77.93±0.67 -> 78.72±0.64 | Helped |
| Add Gaussian augmentation | ProtoNet · NABirds 5-way 1-shot | accuracy: 77.93±0.67 -> 77.94±0.67 | No clear effect |
| Add FinetuneGAN-generated images as data augmentation | ProtoNet · NABirds 5-way 1-shot | accuracy: 77.93±0.67 -> 76.28±0.63 | Hurt |
| Add FinetuneGAN and flip augmentation | MetaIRNet · NABirds 5-way 1-shot | accuracy: 79.21±0.63 -> 79.52±0.62 | Helped |
| Add the diversity loss (ablation) | MFH · AnimalFace | FID: 90.54 -> 75.28 | Helped |
| Add the diversity loss (ablation) | MFH · NABirds | FID: 71.36 -> 42.24 | Helped |
| Use Gumbel-softmax addressing instead of K-means (ablation) | MFH with diversity loss · AnimalFace | FID: 87.63 -> 75.28 | Helped |
| Use Gumbel-softmax addressing instead of K-means (ablation) | MFH with diversity loss · NABirds | FID: 65.72 -> 42.24 | Helped |
Closest papers by meaning
- No Free Lunch: Fundamental Limits of Learning Non-Hallucinating Generative Models (ICLR 2025 · similarity 0.84)
- Analyzing and Mitigating Object Hallucination: A Training Bias Perspective (AAAI 2026 · similarity 0.83)
- Hallucination-aware Intermediate Representation Editing in Large Vision-Lanugage Models (ICLR 2026 · similarity 0.83)
- GHOST: Hallucination-Inducing Image Generation for Multimodal LLMs (ICLR 2026 · similarity 0.83)
- Image Token Matters: Mitigating Hallucination in Discrete Tokenizer-based Large Vision-Language Models via Latent Editing (NeurIPS 2025 · similarity 0.83)
- Self-Correcting Decoding with Generative Feedback for Mitigating Hallucinations in Large Vision-Language Models (ICLR 2025 · similarity 0.83)
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.