Research map / Image generation / Fast diffusion sampling
Map-Guided Caching: A Global Perspective for Efficient Diffusion Transformer
Jiaqi Ji, Ran Yang, Bo Wei, Hui Li, Hee Min Choi
Read the paperCheck your idea against 137,217 papers
AI-extracted summary. Written by AI from the paper's abstract only. Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Proposes Map-Guided Caching, a feature caching framework for Diffusion Transformers that predicts a per-trajectory Temporal Variation Map to decide which intermediate features to cache and reuse.
Problem
Feature caching for accelerating Diffusion Transformers relies on local heuristics or input-agnostic rules to decide which intermediate features to cache and reuse.
Approach
Predicts a Temporal Variation Map (TVMap) for each denoising trajectory and uses it to derive caching decisions globally, instead of local or input-agnostic rules.
Closest papers by meaning
- IdealCache: Rethinking Cache Scheduling in Diffusion Transformers via Ideal Trajectories (NeurIPS 2026 · similarity 0.94)
- HarmoniCa: Harmonizing Training and Inference for Better Feature Caching in Diffusion Transformer Acceleration (ICML 2025 · similarity 0.93)
- Learning-to-Cache: Accelerating Diffusion Transformer via Layer Caching (NeurIPS 2024 · similarity 0.93)
- Accelerating Diffusion Transformers with Token-wise Feature Caching (ICLR 2025 · similarity 0.92)
- BlockDance: Reuse Structurally Similar Spatio-Temporal Features to Accelerate Diffusion Transformers (CVPR 2025 · similarity 0.90)
- Accelerating Diffusion Transformer via Increment-Calibrated Caching with Channel-Aware Singular Value Decomposition (CVPR 2025 · similarity 0.90)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
