Research map / Image generation / Autoregressive and latent generation
Few-shot Generation via Recalling Brain-Inspired Episodic-Semantic Memory
Zhibin Duan, Lv Zhiyi, Chaojie Wang, Bo Chen, Bo An, Mingyuan Zhou
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (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 |
|---|---|---|---|
| add variational structured memory to CNS (VSM-CNS) | CNS · Omniglot-ns | NLL: 89.7 -> 68.2 | Improved |
| add variational structured memory to SCHA-VAE (VSM-SCHA) | SCHA-VAE · Omniglot-ns | NLL: 85.8 -> 56.8 | Improved |
| add semantic memory only to CNS (ablation) | CNS · Omniglot-ns | NLL: 89.7 -> 72.4 | Improved |
| add episodic memory only to CNS (ablation) | CNS · Omniglot-ns | NLL: 89.7 -> 71.8 | Improved |
| use both memories (VSM) rather than only episodic in CNS (ablation) | CNS · Omniglot-ns | NLL: 71.8 -> 68.2 | Improved |
| add semantic memory only to SCHA-VAE (ablation) | SCHA-VAE · Mnist | NLL: 88.5 -> 77.7 | Improved |
| add episodic memory only to SCHA-VAE (ablation) | SCHA-VAE · Double-Mnist | NLL: 53.3 -> 47.2 | Improved |
| add VSM to SCHA-VAE on FS-Cifar100 | SCHA-VAE · FS-Cifar100 | NLL (bpd): 10.338 -> 10.041 | Improved |
| add VSM to FSDM | FSDM · FS-Cifar100 | FID: 35.07 -> 33.21 | Improved |
| add VSM to vFSDM | vFSDM · FS-Cifar100 | FID: 63.73 -> 44.11 | Improved |
| add VSM to vFSDM | vFSDM · Celeba | recall: 0.38 -> 0.40 | Improved |
| add episodic memory alone to vFSDM (ablation) | vFSDM · Celeba | precision: 0.71 -> 0.72 | No clear effect |
Closest papers by meaning
- Learning to Learn Variational Semantic Memory (NeurIPS 2020 · similarity 0.86)
- Task-Free Continual Generation and Representation Learning via Dynamic Expansionable Memory Cluster (AAAI 2024 · similarity 0.86)
- Episodic Memories Generation and Evaluation Benchmark for Large Language Models (ICLR 2025 · similarity 0.85)
- VisMem: Latent Vision Memory Unlocks Potential of Vision-Language Models (CVPR 2026 · similarity 0.85)
- Re-evaluating Continual Learning with Few-Shot Adaptation (NeurIPS 2026 · similarity 0.84)
- IDEA-Bench: How Far are Generative Models from Professional Designing? (CVPR 2025 · similarity 0.84)
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.
