Sparse Forcing adds a native trainable sparsity mechanism and PBSA kernel to autoregressive diffusion video models, yielding higher VBench scores and 1.1-1.27x speedups on 5s to 1min generations.
hub
Spargeattn: Accurate sparse attention accelerating any model inference
22 Pith papers cite this work. Polarity classification is still indexing.
22
Pith papers citing it
hub tools
citation-role summary
background 1
citation-polarity summary
roles
background 1polarities
background 1representative citing papers
Attention sparsity in video DiTs is an input-stable layer-wise property, enabling offline profiling and online bidirectional QK co-clustering for up to 1.93x speedup with PSNR up to 29 dB.
SnapStream deploys sparse KV attention in a production inference system on dataflow accelerators, delivering 4x on-chip memory savings for DeepSeek-671B at 128k context with up to 1832 tokens/sec and minimal accuracy loss on LongBench-v2, AIME24, and LiveCodeBench.
RegimeVGGT applies layer-wise U-shaped compression via saliency-guided banded merging and selectively protected K/V downsampling to deliver 6.7x speedup on VGGT at matched reconstruction quality.
Veda formulates tile selection in video diffusion attention as a reconstruction problem from full attention maps, using statistics-aware and head-aware scoring to enable high sparsity with maintained quality and hardware speedups up to 5.1x end-to-end.
A tiered KV cache architecture computes per-head per-step error bounds on quantized attention and uses adaptive fallback to guarantee bounded or exact outputs relative to FP16 reference.
DashAttention introduces differentiable adaptive sparse hierarchical attention via α-entmax block selection, achieving full-attention accuracy at 75% sparsity with improved Pareto performance over NSA and InfLLMv2.
SparseSAM achieves 2x faster inference and 2.8x memory reduction in SAM with only 0.004 mIoU loss at 0.4 density via Stripe-Sort Attention and Residual-Consistency MLP.
ScaleSearch optimizes block floating point scales via fine-grained search to cut quantization error by 27% for NVFP4, improving PTQ by up to 15 points on MATH500 for Qwen3-8B and attention PPL by 0.77 on Llama 3.1 70B.
AdaCluster delivers a training-free adaptive query-key clustering framework for sparse attention in video DiTs, yielding 1.67-4.31x inference speedup with negligible quality loss on CogVideoX-2B, HunyuanVideo, and Wan-2.1.
BLASST dynamically sparsifies attention by thresholding softmax scores to skip blocks, delivering 1.5x speedups at 70%+ sparsity while preserving benchmark accuracy.
SURF accelerates high-resolution video generation up to 12.5x by using noise reshifting for low-res previews from pretrained models and a shifting-window Refiner for efficient upscaling that retains original signatures.
MTraining scales LLM training to 512K-token contexts on 32 A100 GPUs by integrating dynamic sparse training patterns with balanced and hierarchical sparse ring attention, achieving up to 6x throughput gains without accuracy loss on long-context benchmarks.
Ultra Flash introduces a cascaded streaming super-resolution framework with specialized training, upsampling, and optimization to enable real-time high-resolution video generation from low-res diffusion models.
DVG dynamically selects content-aware spatio-temporal acceleration strategies for diffusion-based video generation, delivering up to 7x speedup with near-lossless quality on models like HunyuanVideo.
Fluxion achieves 1.5x-3.7x speedup in long-context LLM inference with CPU KV caches while limiting accuracy degradation to at most 0.26 relative to full attention.
VFA optimizes Flash Attention by pre-computing global max approximations from key blocks and reordering traversal to reduce vector bottlenecks while preserving exact computation.
DMA delivers a fused low-bit MXFP attention kernel with diagonal tiling that achieves significant speedup on B200 GPUs with negligible generation quality loss.
OSP-Next reports 83.73% VBench score and up to 2.27x speedup via hybrid sparse attention, SSP parallelism, HiF8 quantization, and Mix-GRPO on diffusion transformers.
CORE is a lightweight two-stage prompt compression method for edge-device RAG QA that builds answer and clue sets via NER and semantic matching then refines them to deliver higher accuracy and lower resource costs than baselines.
citing papers explorer
-
Sparse Forcing: Native Trainable Sparse Attention for Real-time Autoregressive Diffusion Video Generation
Sparse Forcing adds a native trainable sparsity mechanism and PBSA kernel to autoregressive diffusion video models, yielding higher VBench scores and 1.1-1.27x speedups on 5s to 1min generations.
-
Attention Sparsity is Input-Stable: Training-Free Sparse Attention for Video Generation via Offline Sparsity Profiling and Online QK Co-Clustering
Attention sparsity in video DiTs is an input-stable layer-wise property, enabling offline profiling and online bidirectional QK co-clustering for up to 1.93x speedup with PSNR up to 29 dB.
-
SnapStream: Efficient Long Sequence Decoding on Dataflow Accelerators
SnapStream deploys sparse KV attention in a production inference system on dataflow accelerators, delivering 4x on-chip memory savings for DeepSeek-671B at 128k context with up to 1832 tokens/sec and minimal accuracy loss on LongBench-v2, AIME24, and LiveCodeBench.
-
RegimeVGGT: Layer-Wise Spatially Preserving Redundancy Removal for Visual Geometry Grounded Transformer
RegimeVGGT applies layer-wise U-shaped compression via saliency-guided banded merging and selectively protected K/V downsampling to deliver 6.7x speedup on VGGT at matched reconstruction quality.
-
Veda: Scalable Video Diffusion via Distilled Sparse Attention
Veda formulates tile selection in video diffusion attention as a reconstruction problem from full attention maps, using statistics-aware and head-aware scoring to enable high sparsity with maintained quality and hardware speedups up to 5.1x end-to-end.
-
Runtime-Certified Bounded-Error Quantized Attention
A tiered KV cache architecture computes per-head per-step error bounds on quantized attention and uses adaptive fallback to guarantee bounded or exact outputs relative to FP16 reference.
-
DashAttention: Differentiable and Adaptive Sparse Hierarchical Attention
DashAttention introduces differentiable adaptive sparse hierarchical attention via α-entmax block selection, achieving full-attention accuracy at 75% sparsity with improved Pareto performance over NSA and InfLLMv2.
-
SparseSAM: Structured Sparsification of Activations in Segment Anything Models
SparseSAM achieves 2x faster inference and 2.8x memory reduction in SAM with only 0.004 mIoU loss at 0.4 density via Stripe-Sort Attention and Residual-Consistency MLP.
-
Search Your Block Floating Point Scales!
ScaleSearch optimizes block floating point scales via fine-grained search to cut quantization error by 27% for NVFP4, improving PTQ by up to 15 points on MATH500 for Qwen3-8B and attention PPL by 0.77 on Llama 3.1 70B.
-
AdaCluster: Adaptive Query-Key Clustering for Sparse Attention in Video Generation
AdaCluster delivers a training-free adaptive query-key clustering framework for sparse attention in video DiTs, yielding 1.67-4.31x inference speedup with negligible quality loss on CogVideoX-2B, HunyuanVideo, and Wan-2.1.
-
BLASST: Dynamic BLocked Attention Sparsity via Softmax Thresholding
BLASST dynamically sparsifies attention by thresholding softmax scores to skip blocks, delivering 1.5x speedups at 70%+ sparsity while preserving benchmark accuracy.
-
SURF: Signature-Retained Fast Video Generation
SURF accelerates high-resolution video generation up to 12.5x by using noise reshifting for low-res previews from pretrained models and a shifting-window Refiner for efficient upscaling that retains original signatures.
-
MTraining: Distributed Dynamic Sparse Attention for Efficient Ultra-Long Context Training
MTraining scales LLM training to 512K-token contexts on 32 A100 GPUs by integrating dynamic sparse training patterns with balanced and hierarchical sparse ring attention, achieving up to 6x throughput gains without accuracy loss on long-context benchmarks.
-
Ultra Flash: Scaling Real-Time Streaming Video Generation to High Resolutions
Ultra Flash introduces a cascaded streaming super-resolution framework with specialized training, upsampling, and optimization to enable real-time high-resolution video generation from low-res diffusion models.
-
Dynamic Video Generation: Shaping Video Generation Across Time and Space
DVG dynamically selects content-aware spatio-temporal acceleration strategies for diffusion-based video generation, delivering up to 7x speedup with near-lossless quality on models like HunyuanVideo.
-
An Efficient Hybrid Sparse Attention with CPU-GPU Parallelism for Long-Context Inference
Fluxion achieves 1.5x-3.7x speedup in long-context LLM inference with CPU KV caches while limiting accuracy degradation to at most 0.26 relative to full attention.
-
VFA: Relieving Vector Operations in Flash Attention with Global Maximum Pre-computation
VFA optimizes Flash Attention by pre-computing global max approximations from key blocks and reordering traversal to reduce vector bottlenecks while preserving exact computation.
-
Diagonal-Tiled Mixed-Precision Attention for Efficient Low-Bit MXFP Inference
DMA delivers a fused low-bit MXFP attention kernel with diagonal tiling that achieves significant speedup on B200 GPUs with negligible generation quality loss.
-
OSP-Next: Efficient High-Quality Video Generation with Sparse Sequence Parallelism, HiF8 Quantization, and Reinforcement Learning
OSP-Next reports 83.73% VBench score and up to 2.27x speedup via hybrid sparse attention, SSP parallelism, HiF8 quantization, and Mix-GRPO on diffusion transformers.
-
Less is More: Lightweight Prompt Compression for Question Answering Applications on Edge Devices
CORE is a lightweight two-stage prompt compression method for edge-device RAG QA that builds answer and clue sets via NER and semantic matching then refines them to deliver higher accuracy and lower resource costs than baselines.
- RT-Lynx: Putting GEMM Sparsity in the Right Place for Diffusion Models
- HEART: Exploiting Head Heterogeneity in Sparse Attention for Video Diffusion