MaskSense: Confronting the Visual Exploration Trap in Masked Image Generation
cs.CV
Submitted: 2026-10-12
Updated: 2026-10-12
Comments: Accepted to NeurIPS 2026. Project Page: https://yawen-shao.github.io/MaskSense/
Code: https://github.com/black-forest-labs/flux
Project page: https://yawen-shao.github.io/MaskSense
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Training Diffusion Models with Reinforcement Learning
- Muse: Text-To-Image Generation via Masked Generative Transformers
- DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
- OpenAI o1 System Card
- Flow Matching for Generative Modeling
- Lumina-mGPT: Illuminate Flexible Photorealistic Text-to-Image Generation with Multimodal Generative Pretraining
- Flow-GRPO: Training Flow Matching Models via Online RL
- Reinforcement Learning Meets Masked Generative Models: Mask-GRPO for Text-to-Image Generation
- Consolidating Reinforcement Learning for Multimodal Discrete Diffusion Models
- Towards Better & Faster Autoregressive Image Generation: From the Perspective of Entropy
- SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis
- Proximal Policy Optimization Algorithms
- Denoising Diffusion Implicit Models
- Wan: Open and Advanced Large-Scale Video Generative Models
- Beyond the 80/20 Rule: High-Entropy Minority Tokens Drive Effective Reinforcement Learning for LLM Reasoning
- Emu3: Next-Token Prediction is All You Need
- Show-o: One Single Transformer to Unify Multimodal Understanding and Generation
- Lumina-DiMOO: An Omni Diffusion Large Language Model for Multi-Modal Generation and Understanding
- DanceGRPO: Unleashing GRPO on Visual Generation
- MMaDA: Multimodal Large Diffusion Language Models
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models