Nepotistically Trained Generative-AI Models Collapse

• URL: http://arxiv.org/abs/2311.12202v2
• Date: Sat, 29 Mar 2025 18:40:09 GMT
• Title: Nepotistically Trained Generative-AI Models Collapse
• Abstract summary: We show that when retrained on even small amounts of their own creation, generative-AI models produce highly distorted images.<n>This distortion extends beyond the text prompts used in retraining, and that once affected, the models struggle to fully heal.
• Score: 13.106063755117399
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Trained on massive amounts of human-generated content, AI-generated image synthesis is capable of reproducing semantically coherent images that match the visual appearance of its training data. We show that when retrained on even small amounts of their own creation, these generative-AI models produce highly distorted images. We also show that this distortion extends beyond the text prompts used in retraining, and that once affected, the models struggle to fully heal even after retraining on only real images.

Related papers

• GMAIL: Generative Modality Alignment for generated Image Learning [51.071351994330605]
  We propose a novel framework for discriminative use of generated images, coined GMAIL, that explicitly treats generated images as a separate modality from real images.<n>Our framework can be easily incorporated with various vision-language models, and we demonstrate its efficacy throughout extensive experiments.
  arXiv  Detail & Related papers  (2026-02-17T05:40:25Z)
• Implicit Neural Representation Facilitates Unified Universal Vision Encoding [11.947746726150001]
  A first-of-its-kind model learns representations which are simultaneously useful for recognition and generation.<n>We train our model as a hyper-network for implicit neural representation, which learns to map images to model weights for fast, accurate reconstruction.<n>The model also learns an unprecedented compressed embedding space with outstanding performance for various visual tasks.
  arXiv  Detail & Related papers  (2026-01-20T18:59:57Z)
• X-Omni: Reinforcement Learning Makes Discrete Autoregressive Image Generative Models Great Again [45.74833463136701]
  We develop a semantic image tokenizer, a unified autoregressive model for both language and images, and an offline diffusion decoder for image generation termed X- Omni.<n>X- Omni achieves state-of-the-art performance in image generation tasks using a 7B language model, producing images with high aesthetic quality while exhibiting strong capabilities in following instructions and rendering long texts.
  arXiv  Detail & Related papers  (2025-07-29T17:59:04Z)
• RealRAG: Retrieval-augmented Realistic Image Generation via Self-reflective Contrastive Learning [54.07026389388881]
  We present the first real-object-based retrieval-augmented generation framework (RealRAG)<n>RealRAG augments fine-grained and unseen novel object generation by learning and retrieving real-world images to overcome the knowledge gaps of generative models.<n>Our framework integrates fine-grained visual knowledge for the generative models, tackling the distortion problem and improving the realism for fine-grained object generation.
  arXiv  Detail & Related papers  (2025-02-02T16:41:54Z)
• Towards Unsupervised Blind Face Restoration using Diffusion Prior [12.69610609088771]
  Blind face restoration methods have shown remarkable performance when trained on large-scale synthetic datasets with supervised learning. These datasets are often generated by simulating low-quality face images with a handcrafted image degradation pipeline. In this paper, we address this issue by using only a set of input images, with unknown degradations and without ground truth targets, to fine-tune a restoration model. Our best model also achieves the state-of-the-art results on both synthetic and real-world datasets.
  arXiv  Detail & Related papers  (2024-10-06T20:38:14Z)
• Data Attribution for Text-to-Image Models by Unlearning Synthesized Images [71.23012718682634]
  The goal of data attribution for text-to-image models is to identify the training images that most influence the generation of a new image. We propose an efficient data attribution method by simulating unlearning the synthesized image. We then identify training images with significant loss deviations after the unlearning process and label these as influential.
  arXiv  Detail & Related papers  (2024-06-13T17:59:44Z)
• AIGCs Confuse AI Too: Investigating and Explaining Synthetic Image-induced Hallucinations in Large Vision-Language Models [37.04195231708092]
  We highlight the exacerbated hallucination phenomena in Large Vision-Language Models (LVLMs) caused by AI-synthetic images. Remarkably, our findings shed light on a consistent AIGC textbfhallucination bias: the object hallucinations induced by synthetic images are characterized by a greater quantity. Our investigations on Q-former and Linear projector reveal that synthetic images may present token deviations after visual projection, thereby amplifying the hallucination bias.
  arXiv  Detail & Related papers  (2024-03-13T13:56:34Z)
• Scaling Laws of Synthetic Images for Model Training ... for Now [54.43596959598466]
  We study the scaling laws of synthetic images generated by state of the art text-to-image models. We observe that synthetic images demonstrate a scaling trend similar to, but slightly less effective than, real images in CLIP training.
  arXiv  Detail & Related papers  (2023-12-07T18:59:59Z)
• Invisible Relevance Bias: Text-Image Retrieval Models Prefer AI-Generated Images [67.18010640829682]
  We show that AI-generated images introduce an invisible relevance bias to text-image retrieval models. The inclusion of AI-generated images in the training data of the retrieval models exacerbates the invisible relevance bias. We propose an effective training method aimed at alleviating the invisible relevance bias.
  arXiv  Detail & Related papers  (2023-11-23T16:22:58Z)
• Image Captions are Natural Prompts for Text-to-Image Models [70.30915140413383]
  We analyze the relationship between the training effect of synthetic data and the synthetic data distribution induced by prompts. We propose a simple yet effective method that prompts text-to-image generative models to synthesize more informative and diverse training data. Our method significantly improves the performance of models trained on synthetic training data.
  arXiv  Detail & Related papers  (2023-07-17T14:38:11Z)
• Hybrid Neural Rendering for Large-Scale Scenes with Motion Blur [68.24599239479326]
  We develop a hybrid neural rendering model that makes image-based representation and neural 3D representation join forces to render high-quality, view-consistent images. Our model surpasses state-of-the-art point-based methods for novel view synthesis.
  arXiv  Detail & Related papers  (2023-04-25T08:36:33Z)
• Synthetic Image Data for Deep Learning [0.294944680995069]
  Realistic synthetic image data rendered from 3D models can be used to augment image sets and train image classification semantic segmentation models. We show how high quality physically-based rendering and domain randomization can efficiently create a large synthetic dataset based on production 3D CAD models of a real vehicle.
  arXiv  Detail & Related papers  (2022-12-12T20:28:13Z)
• Is synthetic data from generative models ready for image recognition? [69.42645602062024]
  We study whether and how synthetic images generated from state-of-the-art text-to-image generation models can be used for image recognition tasks. We showcase the powerfulness and shortcomings of synthetic data from existing generative models, and propose strategies for better applying synthetic data for recognition tasks.
  arXiv  Detail & Related papers  (2022-10-14T06:54:24Z)
• From Face to Natural Image: Learning Real Degradation for Blind Image Super-Resolution [72.68156760273578]
  We design training pairs for super-resolving the real-world low-quality (LQ) images. We take paired HQ and LQ face images as inputs to explicitly predict degradation-aware and content-independent representations. We then transfer these real degradation representations from face to natural images to synthesize the degraded LQ natural images.
  arXiv  Detail & Related papers  (2022-10-03T08:09:21Z)
• Synthetic-to-Real Domain Adaptation using Contrastive Unpaired Translation [28.19031441659854]
  We propose a multi-step method to obtain training data without manual annotation effort. From 3D object meshes, we generate images using a modern synthesis pipeline. We utilize a state-of-the-art image-to-image translation method to adapt the synthetic images to the real domain.
  arXiv  Detail & Related papers  (2022-03-17T17:13:23Z)
• Counterfactual Generative Networks [59.080843365828756]
  We propose to decompose the image generation process into independent causal mechanisms that we train without direct supervision. By exploiting appropriate inductive biases, these mechanisms disentangle object shape, object texture, and background. We show that the counterfactual images can improve out-of-distribution with a marginal drop in performance on the original classification task.
  arXiv  Detail & Related papers  (2021-01-15T10:23:12Z)
• Blind Motion Deblurring through SinGAN Architecture [21.104218472462907]
  Blind motion deblurring involves reconstructing a sharp image from an observation that is blurry. SinGAN is a generative model that is unconditional and could be learned from a single natural image.
  arXiv  Detail & Related papers  (2020-11-07T06:09:16Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.