model efficiency

8 papers

• CLIPGaussian: Universal and Multimodal Style Transfer Based on Gaussian Splatting
• Native Segmentation Vision Transformers
• Neural Entropy
• OmniDraft: A cross-vocabulary, online adaptive drafter for on-device speculative decoding
• PocketSR: The Super-Resolution Expert in Your Pocket Mobiles
• SkyLadder: Better and Faster Pretraining via Context Window Scheduling
• The Rise of Parameter Specialization for Knowledge Storage in Large Language Models
• Toward Efficient Inference Attacks: Shadow Model Sharing via Mixture-of-Experts