paper-with-me

Papers

Ditto: Accelerating Diffusion Model via Temporal Value Similarity

2025-01-20 · Sungbin Kim, Hyunwuk Lee, Wonho Cho, Mincheol Park, Won Woo Ro

Diffusion models achieve superior performance in image generation tasks. However, it incurs significant computation overheads due to its iterative structure. To address these overheads, we analyze this iterative structure and observe that adjacent time steps in diffusion models exhibit high value similarity, leading to narrower differences between consecutive time steps. We adapt these characteristics to a quantized diffusion model and reveal that the majority of these differences can be represented with reduced bit-width, and even zero. Based on our observations, we propose the Ditto algorithm, a difference processing algorithm that leverages temporal similarity with quantization to enhance the efficiency of diffusion models. By exploiting the narrower differences and the distributive property of layer operations, it performs full bit-width operations for the initial time step and processes subsequent steps with temporal differences. In addition, Ditto execution flow optimization is designed to mitigate the memory overhead of temporal difference processing, further boosting the efficiency of the Ditto algorithm. We also design the Ditto hardware, a specialized hardware accelerator, fully exploiting the dynamic characteristics of the proposed algorithm. As a result, the Ditto hardware achieves up to 1.5x speedup and 17.74% energy saving compared to other accelerators.

📄 PDF Abstract BibTeX arXiv:2501.11211

Code (0)

등록된 구현이 없습니다.

Tasks

Image GenerationmodelQuantization

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…

Similar Papers 제목 키워드 기반

A Diffusion-Based Framework for Configurable and Realistic Multi-Storage Trace Generation

2025-09-02 · Seohyun Kim, Junyoung Lee, Jongho Park, Jinhyung Koo 외 arxiv

We propose DiTTO, a novel diffusion-based framework for generating realistic, precisely configurable, and diverse multi-device storage traces. Leveraging advanced diffusion techniques, DiTTO enables the synthesis of high…

Real-time Inference and Extrapolation via a Diffusion-inspired Temporal Transformer Operator (DiTTO)

2023-07-18 · Oded Ovadia, Vivek Oommen, Adar Kahana, Ahmad Peyvan 외

Extrapolation remains a grand challenge in deep neural networks across all application domains. We propose an operator learning method to solve time-dependent partial differential equations (PDEs) continuously and with e…

Operator learningSuper-Resolution

DiTTo-TTS: Diffusion Transformers for Scalable Text-to-Speech without Domain-Specific Factors

2024-06-17 · Keon Lee, Dong Won Kim, Jaehyeon Kim, Seungjun Chung 외

Large-scale latent diffusion models (LDMs) excel in content generation across various modalities, but their reliance on phonemes and durations in text-to-speech (TTS) limits scalability and access from other fields. Whil…

text-to-speechText to Speech

DITTO: Diffusion Inference-Time T-Optimization for Music Generation

2024-01-22 · Zachary Novack, Julian McAuley, Taylor Berg-Kirkpatrick, Nicholas J. Bryan

We propose Diffusion Inference-Time T-Optimization (DITTO), a general-purpose frame-work for controlling pre-trained text-to-music diffusion models at inference-time via optimizing initial noise latents. Our method can b…

Computational EfficiencyMusic Generation

DITTO-2: Distilled Diffusion Inference-Time T-Optimization for Music Generation

2024-05-30 · Zachary Novack, Julian McAuley, Taylor Berg-Kirkpatrick, Nicholas Bryan

Controllable music generation methods are critical for human-centered AI-based music creation, but are currently limited by speed, quality, and control design trade-offs. Diffusion Inference-Time T-optimization (DITTO), …

Music Generation