Stars
Ambiguous Handwritten Mathematical Expression Generator
Unsupervised Training Data Generation of Handwritten Formulas using Generative Adversarial Networks with Self-Attention
[ICLR'25 Oral] Representation Alignment for Generation: Training Diffusion Transformers Is Easier Than You Think
[ICML'26] Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
This is the largest dataset on the evolution of Chinese characters, containing characters from multiple periods including oracle bone script, bronze inscriptions, Spring and Autumn period character…
Repository for "Graph Flow Matching: Enhancing Image Generation with Neighbor-Aware Flow Fields"
Chronicles-OCR: A Cross-Temporal Perception Benchmark for the Evolutionary Trajectory of Chinese Characters (Seven Chinese Scripts, 2800 images)
[ICML 2024] LESS: Selecting Influential Data for Targeted Instruction Tuning
[arXiv 2026] DiffMath: Symbol- and Graph-Aware Latent Diffusion Transformer for Handwritten Mathematical Expression Generation
open source style transfer model on par with nano banana pro, supporting Qwen-Image-Edit 2509, 2511, SenseNovaU1
Official PyTorch re-implementation of MiniT2I.
Toy-scale unified multimodal model experiments — encoder-free understanding & generation with Mixture-of-Transformers on MLX/Apple Silicon
[Innovation 2026] Oracle bone script decipherment via human-workflow-inspired deep learning
Latex template for ACM conference/Journal rebuttal.
Official PyTorch Implementation of "P-HTG: One-Shot Handwritten Text Generation via Prototype-Guided Adaptive Gated Fusion"
[ACMMM'2024] Generative Expressive Conversational Speech Synthesis
The official implementation of "Frequency-Aware Flow Matching for High-Quality Image Generation"
(ෆ`꒳´ෆ) A Survey on Text-to-Image Generation/Synthesis.
A curated list of resources related to linear attention mechanisms.
⚡ Clash for Lab 是为实验室环境设计的科学上网工具，无需sudo权限，优雅地一键式脚本安装
[ECCV 2026] Official repository of FlowInOne: Unifying Multimodal Generation as Image-In Image-Out Flow Matching
[ICLR 2026 Oral] Locality-aware Parallel Decoding for Efficient Autoregressive Image Generation
FineStyle: Fine-grained Controllable Style Personalization for Text-to-image Models