Cover Image

SpeCa: Accelerating Diffusion Transformers with Speculative Feature Caching

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Liu, Jiacheng, Zou, Chang, Lyu, Yuanhuiyi, Ren, Fei, Wang, Shaobo, Li, Kaixin, Zhang, Linfeng
Format:	Preprint
Published:	2025
Subjects:	Machine Learning Artificial Intelligence Computer Vision and Pattern Recognition
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

From Reusing to Forecasting: Accelerating Diffusion Models with TaylorSeers
by: Liu, Jiacheng, et al.
Published: (2025)

Accelerating Diffusion Transformers with Token-wise Feature Caching
by: Zou, Chang, et al.
Published: (2024)

DisCa: Accelerating Video Diffusion Transformers with Distillation-Compatible Learnable Feature Caching
by: Zou, Chang, et al.
Published: (2026)

Rethinking Token-wise Feature Caching: Accelerating Diffusion Transformers with Dual Feature Caching
by: Zou, Chang, et al.
Published: (2024)

FreqCa: Accelerating Diffusion Models via Frequency-Aware Caching
by: Liu, Jiacheng, et al.
Published: (2025)

Compute Only 16 Tokens in One Timestep: Accelerating Diffusion Transformers with Cluster-Driven Feature Caching
by: Zheng, Zhixin, et al.
Published: (2025)

Let Features Decide Their Own Solvers: Hybrid Feature Caching for Diffusion Transformers
by: Zheng, Shikang, et al.
Published: (2025)

Not All Samples Should Be Utilized Equally: Towards Understanding and Improving Dataset Distillation
by: Wang, Shaobo, et al.
Published: (2024)

CacheQuant: Comprehensively Accelerated Diffusion Models
by: Liu, Xuewen, et al.
Published: (2025)

BWCache: Accelerating Video Diffusion Transformers through Block-Wise Caching
by: Cui, Hanshuai, et al.
Published: (2025)

Forecast then Calibrate: Feature Caching as ODE for Efficient Diffusion Transformers
by: Zheng, Shikang, et al.
Published: (2025)

Accelerating Diffusion-based Video Editing via Heterogeneous Caching: Beyond Full Computing at Sampled Denoising Timestep
by: Liu, Tianyi, et al.
Published: (2026)

A Survey on Cache Methods in Diffusion Models: Toward Efficient Multi-Modal Generation
by: Liu, Jiacheng, et al.
Published: (2025)

Accelerating Diffusion Models with One-to-Many Knowledge Distillation
by: Zhang, Linfeng, et al.
Published: (2024)

AdaCorrection: Adaptive Offset Cache Correction for Accurate Diffusion Transformers
by: Liu, Dong, et al.
Published: (2026)

LeMiCa: Lexicographic Minimax Path Caching for Efficient Diffusion-Based Video Generation
by: Gao, Huanlin, et al.
Published: (2025)

HarmoniCa: Harmonizing Training and Inference for Better Feature Caching in Diffusion Transformer Acceleration
by: Huang, Yushi, et al.
Published: (2024)

H2-Cache: A Novel Hierarchical Dual-Stage Cache for High-Performance Acceleration of Generative Diffusion Models
by: Sung, Mingyu, et al.
Published: (2025)

HiCache: A Plug-in Scaled-Hermite Upgrade for Taylor-Style Cache-then-Forecast Diffusion Acceleration
by: Feng, Liang, et al.
Published: (2025)

Unified Batch Normalization: Identifying and Alleviating the Feature Condensation in Batch Normalization and a Unified Framework
by: Wang, Shaobo, et al.
Published: (2023)

Towards Principled Dataset Distillation: A Spectral Distribution Perspective
by: Wu, Ruixi, et al.
Published: (2026)

LANTERN: Accelerating Visual Autoregressive Models with Relaxed Speculative Decoding
by: Jang, Doohyuk, et al.
Published: (2024)

UNSEEN: Enhancing Dataset Pruning from a Generalization Perspective
by: Xu, Furui, et al.
Published: (2025)

CTA-Net: A CNN-Transformer Aggregation Network for Improving Multi-Scale Feature Extraction
by: Meng, Chunlei, et al.
Published: (2024)

FastCache: Fast Caching for Diffusion Transformer Through Learnable Linear Approximation
by: Liu, Dong, et al.
Published: (2025)

FRDiff : Feature Reuse for Universal Training-free Acceleration of Diffusion Models
by: So, Junhyuk, et al.
Published: (2023)

Generative Pre-trained Autoregressive Diffusion Transformer
by: Zhang, Yuan, et al.
Published: (2025)

HIPPO: Accelerating Video Large Language Models Inference via Holistic-aware Parallel Speculative Decoding
by: Lv, Qitan, et al.
Published: (2026)

SpecVLM: Fast Speculative Decoding in Vision-Language Models
by: Huang, Haiduo, et al.
Published: (2025)

Focused Forcing: Content-Aware Per-Frame KV Selection for Efficient Autoregressive Video Diffusion
by: Cai, Peiliang, et al.
Published: (2026)

OmniCache: A Trajectory-Oriented Global Perspective on Training-Free Cache Reuse for Diffusion Transformer Models
by: Chu, Huanpeng, et al.
Published: (2025)

VVS: Accelerating Speculative Decoding for Visual Autoregressive Generation via Partial Verification Skipping
by: Dong, Haotian, et al.
Published: (2025)

Socratic-Geo: Synthetic Data Generation and Geometric Reasoning via Multi-Agent Interaction
by: Jiao, Zhengbo, et al.
Published: (2026)

AirCache: Activating Inter-modal Relevancy KV Cache Compression for Efficient Large Vision-Language Model Inference
by: Huang, Kai, et al.
Published: (2025)

Decouple-Then-Merge: Finetune Diffusion Models as Multi-Task Learning
by: Ma, Qianli, et al.
Published: (2024)

Chipmunk: Training-Free Acceleration of Diffusion Transformers with Dynamic Column-Sparse Deltas
by: Silveria, Austin, et al.
Published: (2025)

Relational Feature Caching for Accelerating Diffusion Transformers
by: Son, Byunggwan, et al.
Published: (2026)

Motion-Aware Caching for Efficient Autoregressive Video Generation
by: Xu, Jing, et al.
Published: (2026)

Forecast the Principal, Stabilize the Residual: Subspace-Aware Feature Caching for Efficient Diffusion Transformers
by: Chen, Guantao, et al.
Published: (2026)

STaR-KV: Spatio-Temporal Adaptive Re-weighting for KV Cache Compression in GUI Vision-Language Models
by: Han, Yuhang, et al.
Published: (2026)