paper-with-me

홈 › Papers

Understanding the Quality-Diversity Trade-off in Diffusion Language Models

2025-03-11 · Zak Buzzard

Diffusion models have seen immense success in modelling continuous data across a range of domains such as vision and audio. Despite the challenges of adapting diffusion models to discrete data, recent work explores their application to text generation by working in the continuous embedding space. However, these models lack a natural means to control the inherent trade-off between quality and diversity as afforded by the temperature hyperparameter in autoregressive models, hindering understanding of model performance and restricting generation quality. This work proposes the use of classifier-free guidance and stochastic clamping for manipulating the quality-diversity trade-off on sequence-to-sequence tasks, demonstrating that these techniques may be used to improve the performance of a diffusion language model.

📄 PDF Abstract BibTeX arXiv:2503.10683

Code (1)

zzbuzzard/guidediffuseq 공식 구현 pytorch

Tasks

DiversityLanguage ModelingLanguage ModellingText 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

DiverseVAR: Balancing Diversity and Quality of Next-Scale Visual Autoregressive Models

2025-11-26 · Mingue Park, Prin Phunyaphibarn, Phillip Y. Lee, Minhyuk Sung arxiv

We introduce DiverseVAR, a framework that enhances the diversity of text-conditioned visual autoregressive models (VAR) at test time without requiring retraining, fine-tuning, or substantial computational overhead. While…

Image Generation

BIRDS: Characterizing and Understanding Biodiversity Impact of Large Language Model Serving

2026-05-26 · Tianyao Shi, Yi Ding arxiv

Large language model (LLM) serving creates environmental impacts beyond carbon and water, including ecosystem damage through biodiversity-related pathways. We present BIRDS, a framework for Biodiversity Impact of Request…

Elucidating Optimal Reward-Diversity Tradeoffs in Text-to-Image Diffusion Models

2024-09-09 · Rohit Jena, Ali Taghibakhshi, Sahil Jain, Gerald Shen 외

Text-to-image (T2I) diffusion models have become prominent tools for generating high-fidelity images from text prompts. However, when trained on unfiltered internet data, these models can produce unsafe, incorrect, or st…

Diversity

Text-to-Sticker: Style Tailoring Latent Diffusion Models for Human Expression

2023-11-17 · Animesh Sinha, Bo Sun, Anmol Kalia, Arantxa Casanova 외

We introduce Style Tailoring, a recipe to finetune Latent Diffusion Models (LDMs) in a distinct domain with high visual quality, prompt alignment and scene diversity. We choose sticker image generation as the target doma…

DiversityImage GenerationPrompt Engineering

SwiftBrush v2: Make Your One-step Diffusion Model Better Than Its Teacher

2024-08-26 · Trung Dao, Thuan Hoang Nguyen, Thanh Le, Duc Vu 외

In this paper, we aim to enhance the performance of SwiftBrush, a prominent one-step text-to-image diffusion model, to be competitive with its multi-step Stable Diffusion counterpart. Initially, we explore the quality-di…

Diversity