EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts Paper • 2606.18967 • Published 15 days ago • 24
XQuant: Breaking the Memory Wall for LLM Inference with KV Cache Rematerialization Paper • 2508.10395 • Published Aug 14, 2025 • 42
UNCAGE: Contrastive Attention Guidance for Masked Generative Transformers in Text-to-Image Generation Paper • 2508.05399 • Published Aug 7, 2025 • 17