Circumpolar Wall-to-Wall Detection of Retrogressive Thaw Slumps with Scarce Labels and Extreme Class Imbalance
cs.CV
Submitted: 2026-10-09
Updated: 2026-10-09
Code: https://github.com/whrc/RTSmapping_v2
Terminology
Sources
- AlphaEarth Foundations: An embedding field model for accurate and efficient global mapping from sparse label data
- On Calibration of Modern Neural Networks
- Averaging Weights Leads to Wider Optima and Better Generalization
- SGDR: Stochastic Gradient Descent with Warm Restarts
- Decoupled Weight Decay Regularization
- Mixed Precision Training
- Calibrating Deep Neural Networks using Focal Loss
- TrivialAugment: Tuning-free Yet State-of-the-Art Data Augmentation
- SAM 2: Segment Anything in Images and Videos
- DINOv3
- EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs