J C's picture

J C

dark-pen

·

AI & ML interests

None yet

Recent Activity

upvoted a paper about 1 hour ago

Enhanced SCanNet with CBAM and Dice Loss for Semantic Change Detection

liked a dataset about 1 hour ago

BEE-spoke-data/awesome-python-apps

reacted to buddhi19's post with 🚀 about 1 hour ago

Article Highlight: SyntheticGen, Controllable Diffusion for Long-Tail Remote Sensing 🛰️ Why is remote-sensing segmentation still hard—even with strong models? Because the issue is not only the model… it’s the data. In real-world datasets like LoveDA, class distributions are highly imbalanced, and the problem is compounded by Urban/Rural domain shifts, where visual characteristics and class frequencies differ significantly. This leads to poor learning for minority classes and weak generalization. ⚖️ The Idea: Make Data Controllable Instead of treating data augmentation as a random process, SyntheticGen turns it into a controllable pipeline. 👉 What if you could: Specify which classes you want more of? Control how much of each class appears? Generate data that respects domain (Urban/Rural) characteristics? That’s exactly what SyntheticGen enables. 🧠 How It Works SyntheticGen introduces a structured generation process: Layout Generation (Stage A) A ratio-conditioned discrete diffusion model generates semantic layouts that match user-defined class distributions. Image Synthesis (Stage B) A ControlNet-guided Stable Diffusion pipeline converts layouts into realistic remote-sensing imagery. 💡 This separation between semantic control and visual realism is key—it allows both precision and high-quality generation. Why It Matters Tackles long-tail imbalance directly at the data level Improves minority-class segmentation performance Enhances cross-domain generalization (Urban ↔ Rural) Moves toward data-centric AI, where we design training data—not just models Recent research shows that diffusion-based synthetic data can significantly improve performance in long-tailed settings by generating high-value samples for rare or difficult cases . SyntheticGen takes this further by making the process explicitly controllable, not just generative. 📄 Paper https://arxiv.org/abs/2602.04749 💻 Code & Synthetic Data https://github.com/Buddhi19/SyntheticGen

View all activity

Organizations

Collections 7

View 7 collections

models 15

dark-pen/ExGRPO-LUFFY-7B-Continual-IQ4_XS-GGUF

Text Generation • 8B • Updated 20 days ago • 194

dark-pen/Remix-R1-Distilled-Qwen-7B-IQ4_XS-GGUF

8B • Updated 20 days ago • 193

dark-pen/Klear-AgentForge-8B-IQ4_XS-GGUF

8B • Updated 20 days ago • 205

dark-pen/Eva-4B-V2-IQ4_XS-GGUF

Text Classification • 4B • Updated Feb 9 • 3 • 2

dark-pen/EmbeddingsG300M-ft-Q6_K-GGUF

Sentence Similarity • 0.3B • Updated Jan 23 • 5

dark-pen/DASD-4B-Thinking-IQ4_XS-GGUF

Text Generation • 4B • Updated Jan 17 • 9

dark-pen/mR3-Qwen3-4B-en-prompt-en-thinking-IQ4_XS-GGUF

4B • Updated Jan 14 • 25

dark-pen/Nemotron-Orchestrator-8B-IQ4_XS-GGUF

8B • Updated Jan 9 • 9

dark-pen/Logics-STEM-8B-RL-IQ4_XS-GGUF

8B • Updated Jan 9 • 23

dark-pen/brief-pro-IQ4_XS-GGUF

4B • Updated Dec 15, 2025 • 2

datasets 0

None public yet