generative capabilities
Generative capabilities pertain to the ability of models to create new data samples that resemble a training dataset, applicable in tasks like image generation, text synthesis, and style transfer.
- A Diffusion Model for Regular Time Series Generation from Irregular Data with Completion and Masking
- Adaptive Inference-Time Scaling via Cyclic Diffusion Search
- Adjusting Initial Noise to Mitigate Memorization in Text-to-Image Diffusion Models
- Hybrid Latent Reasoning via Reinforcement Learning
- Incentivizing Truthful Language Models via Peer Elicitation Games
- MMCSBench: A Fine-Grained Benchmark for Large Vision-Language Models in Camouflage Scenes
- Semantic Surgery: Zero-Shot Concept Erasure in Diffusion Models
- Unified Reinforcement and Imitation Learning for Vision-Language Models
- WISA: World simulator assistant for physics-aware text-to-video generation