DeCoPrune: Efficient KV-Cache Pruning for Autoregressive Video Diffusion via Denoising Consistency

arXiv:2609.39096 · cs.CV · Submitted 2026-09-30 · Read on arXiv

cs.CV

Submitted: 2026-09-30

Updated: 2026-10-05

Code: https://github.com/DeCoPrune/CMBench

Project page: https://decoprune.github.io

Terminology

• Consistency — 40× in this paper · explained in 3 episodes
• Pruning — 37× in this paper · explained in 4 episodes
• Benchmark — 13× in this paper · explained in 2 episodes
• World Model — 11× in this paper · explained in 7 episodes
• KV cache — 10× in this paper · explained in 3 episodes
• Diffusion Model — 9× in this paper · explained in 8 episodes
• Diffusion Models — 7× in this paper · explained in 6 episodes
• Transformer — 6× in this paper · explained in 2 episodes
• Ablation — 4× in this paper · explained in 1 episode
• Distillation — 4× in this paper · explained in 2 episodes
• Redundancy — 4× in this paper · explained in 2 episodes
• Embeddings — 3× in this paper · explained in 2 episodes
• Sparse Attention — 3× in this paper · explained in 5 episodes
• Aggregation — 2× in this paper · explained in 1 episode
• Benchmarking — 2× in this paper · explained in 3 episodes
• Calibration — 2× in this paper · explained in 7 episodes
• Core — 2× in this paper · explained in 2 episodes
• Fidelity — 2× in this paper · explained in 2 episodes
• Generative Model — 2× in this paper · explained in 2 episodes
• Generative Models — 2× in this paper · explained in 2 episodes
• Ablation study — 1× in this paper · explained in 1 episode
• Accuracy — 1× in this paper · explained in 1 episode
• Adam — 1× in this paper · explained in 2 episodes
• Addressability — 1× in this paper · explained in 1 episode
• Distribution Matching — 1× in this paper · explained in 2 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Quantization — 1× in this paper · explained in 12 episodes
• RAG — 1× in this paper · explained in 3 episodes
• Sparsity — 1× in this paper · explained in 4 episodes
• Temporal consistency — 1× in this paper · explained in 2 episodes

Sources

• Diffusion for World Modeling: Visual Details Matter in Atari
• Focused Forcing: Content-Aware Per-Frame KV Selection for Efficient Autoregressive Video Diffusion
• PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
• Diffusion Forcing: Next-token Prediction Meets Full-Sequence Diffusion
• SkyReels-V2: Infinite-length Film Generative Model
• Past- and Future-Informed KV Cache Policy with Salience Estimation in Autoregressive Video Diffusion
• MemoBench: Benchmarking World Modeling in Dynamically Changing Environments
• Pyramid Forcing: Head-Aware Pyramid KV Cache Policy for High-Quality Long Video Generation
• VRAG: Learning World Models for Interactive Video Generation
• Self-Forcing++: Towards Minute-Scale High-Quality Video Generation
• SlotMemory: Object-Centric KV Memory for Streaming Long-Video Generation
• NarrLV: Towards a Comprehensive Narrative-Centric Evaluation for Long Video Generation
• Infinite Worlds with Versatile Interactions
• Efficient Autoregressive Video Diffusion with Dummy Head
• MineWorld: a Real-Time and Open-Source Interactive World Model on Minecraft
• Matrix-game 2.0: An open-source, real-time, and streaming interactive world model
• LongLive-RAG: A General Retrieval-Augmented Framework for Long Video Generation
• Self Forcing: Bridging the Train-Test Gap in Autoregressive Video Diffusion
• VBench: Comprehensive Benchmark Suite for Video Generative Models

Related papers

• Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
• AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
• Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
• MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
• TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
• A Survey on Efficient Vision-Language-Action Models