paper-with-me

홈 › Papers

Generating on Generated: An Approach Towards Self-Evolving Diffusion Models

2025-02-14 · Xulu Zhang, XiaoYong Wei, Jinlin Wu, Jiaxin Wu, Zhaoxiang Zhang, Zhen Lei, Qing Li

Recursive Self-Improvement (RSI) enables intelligence systems to autonomously refine their capabilities. This paper explores the application of RSI in text-to-image diffusion models, addressing the challenge of training collapse caused by synthetic data. We identify two key factors contributing to this collapse: the lack of perceptual alignment and the accumulation of generative hallucinations. To mitigate these issues, we propose three strategies: (1) a prompt construction and filtering pipeline designed to facilitate the generation of perceptual aligned data, (2) a preference sampling method to identify human-preferred samples and filter out generative hallucinations, and (3) a distribution-based weighting scheme to penalize selected samples with hallucinatory errors. Our extensive experiments validate the effectiveness of these approaches.

📄 PDF Abstract BibTeX arXiv:2502.09963

Code (0)

등록된 구현이 없습니다.

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

Diffusion Curriculum: Synthetic-to-Real Generative Curriculum Learning via Image-Guided Diffusion

2024-10-17 · Yijun Liang, Shweta Bhardwaj, Tianyi Zhou

Low-quality or scarce data has posed significant challenges for training deep neural networks in practice. While classical data augmentation cannot contribute very different new data, diffusion models opens up a new door…

Data AugmentationImage GenerationSelf-Evolving AI

Confidence-Orchestrated Self-Evolution against Uncertain LLM Feedback

2026-05-27 · Bowen Wei, Nan Wang, Yuqing Zhou, Jinhao Pan 외 arxiv

Self-evolving large language models (LLMs) learn by generating their own training tasks and solutions, reducing reliance on human-curated supervision. However, in many reasoning domains, the model must also validate gene…

DiffusionFace: Towards a Comprehensive Dataset for Diffusion-Based Face Forgery Analysis

2024-03-27 · Zhongxi Chen, Ke Sun, Ziyin Zhou, Xianming Lin 외

The rapid progress in deep learning has given rise to hyper-realistic facial forgery methods, leading to concerns related to misinformation and security risks. Existing face forgery datasets have limitations in generatin…

Image GenerationMisinformation

Attentive Eraser: Unleashing Diffusion Model's Object Removal Potential via Self-Attention Redirection Guidance

2024-12-17 · Wenhao Sun, Benlei Cui, Xue-Mei Dong, Jingqun Tang

Recently, diffusion models have emerged as promising newcomers in the field of generative models, shining brightly in image generation. However, when employed for object removal tasks, they still encounter issues such as…

Image GenerationObject

ANCHOR: An External LLM-Driven Supervisory Module Facilitating Healthy Evolution in Self-Evolving Systems

2026-06-04 · Dianxing Shi, Bowen Wang, Junqi He, Junhao Chen 외 arxiv

Self-evolving agents improve through continual self-play and self-generated learning signals, but their internally generated tasks and verifier signals provide limited coverage of phase-level errors, allowing capability …

Mathematical Reasoning