paper-with-me

Papers

Soft Truncation: A Universal Training Technique of Score-based Diffusion Model for High Precision Score Estimation

2021-06-10 · Dongjun Kim, Seungjae Shin, Kyungwoo Song, Wanmo Kang, Il-Chul Moon

Recent advances in diffusion models bring state-of-the-art performance on image generation tasks. However, empirical results from previous research in diffusion models imply an inverse correlation between density estimation and sample generation performances. This paper investigates with sufficient empirical evidence that such inverse correlation happens because density estimation is significantly contributed by small diffusion time, whereas sample generation mainly depends on large diffusion time. However, training a score network well across the entire diffusion time is demanding because the loss scale is significantly imbalanced at each diffusion time. For successful training, therefore, we introduce Soft Truncation, a universally applicable training technique for diffusion models, that softens the fixed and static truncation hyperparameter into a random variable. In experiments, Soft Truncation achieves state-of-the-art performance on CIFAR-10, CelebA, CelebA-HQ 256x256, and STL-10 datasets.

📄 PDF Abstract BibTeX arXiv:2106.05527

Code (1)

Kim-Dongjun/Soft-Truncation 공식 구현 pytorch

Tasks

Density EstimationImage Generation

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

Breaking the Top-$K$ Barrier: Advancing Top-$K$ Ranking Metrics Optimization in Recommender Systems

2025-08-04 · Weiqin Yang, Jiawei Chen, Shengjia Zhang, Peng Wu 외 arxiv

In the realm of recommender systems (RS), Top-$K$ ranking metrics such as NDCG@$K$ are the gold standard for evaluating recommendation performance. However, during the training of recommendation models, optimizing NDCG@$…

Computational Efficiency

Sparse Text Generation

2020-04-06 · EMNLP 2020 11 · Pedro Henrique Martins, Zita Marinho, André F. T. Martins

Current state-of-the-art text generators build on powerful language models such as GPT-2, achieving impressive performance. However, to avoid degenerate text, they require sampling from a modified softmax, via temperatur…

Dialogue GenerationDiversityLanguage ModelingLanguage Modelling+2

Universal Approximation with Softmax Attention

2025-04-22 · Jerry Yao-Chieh Hu, Hude Liu, Hong-Yu Chen, Weimin Wu 외

We prove that with linear transformations, both (i) two-layer self-attention and (ii) one-layer self-attention followed by a softmax function are universal approximators for continuous sequence-to-sequence functions on c…

EntmaxKV: Support-Aware Decoding for Entmax Attention

2026-05-20 · Gonçalo Duarte, Miguel Couceiro, Marcos V. Treviso arxiv

Long-context decoding is increasingly limited by KV-cache memory traffic since each generated token attends over a cache whose size grows linearly with context length. Existing sparse decoding methods reduce this cost by…

Multi-level Latent Space Structuring for Generative Control

2022-02-11 · Oren Katzir, Vicky Perepelook, Dani Lischinski, Daniel Cohen-Or

Truncation is widely used in generative models for improving the quality of the generated samples, at the expense of reducing their diversity. We propose to leverage the StyleGAN generative architecture to devise a new t…

Diversity