Can We Predict The Human Preference For Text-to-Image Content Prior To Generation And Is It Even Useful To Do So?

• URL: http://arxiv.org/abs/2606.05478v1
• Date: Wed, 03 Jun 2026 21:57:05 GMT
• Title: Can We Predict The Human Preference For Text-to-Image Content Prior To Generation And Is It Even Useful To Do So?
• Abstract summary: Diffusion Models (DM) have revolutionized text-driven generation by enabling the synthesis of high-quality, photorealistic visual content from user prompts.<n>We investigate what extent we can predict scalar Human Preference Metrics (HPM) scores prior to committing compute resources for generation.<n>Our investigation reveals that not only is this possible, but that it is feasible to achieve negligible hardware overhead.
• Score: 4.8986598953553555
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Diffusion Models (DM) have revolutionized text-driven generation by enabling the synthesis of high-quality, photorealistic visual content from user prompts. Whereas prior advances in visual generation such as VAEs and GANs were primarily evaluated on perceptual or visual similarity metrics such as FID PSNR, DM advances have fostered the development of more advanced Human Preference Metrics (HPM) that model and quantify human judgment as scalar values. However, DMs synthesize content using an inherently stochastic process where random noise seeds generation. The initial random noise directly affects the quality of generated outputs, both qualitatively and quantitatively. This influence is pronounced in smaller models for local deployment scenarios. Given this phenomenon, we first investigate to what extent we can predict scalar HPM scores prior to committing compute resources for generation. Further, we then investigate to what extent we can leverage such prediction to improve the quality of generated images, and also study which HPMs are best suited for this task. Our investigation reveals that not only is this possible, but that it is feasible to achieve negligible hardware overhead.

Related papers

• VLM-Guided Group Preference Alignment for Diffusion-based Human Mesh Recovery [75.62565146049015]
  We introduce a dual-memory augmented HMR critique agent with self-reflection to produce context-aware quality scores for predicted meshes.<n>These scores distill fine-grained cues about 3D human motion structure, physical feasibility, and alignment with the input image.<n>We propose a group preference alignment framework for finetuning diffusion-based HMR models.
  arXiv  Detail & Related papers  (2026-02-22T13:19:06Z)
• More Than Generation: Unifying Generation and Depth Estimation via Text-to-Image Diffusion Models [53.98725993420285]
  Generative depth estimation methods leverage the rich visual priors stored in pre-trained text-to-image diffusion models.<n>We introduce MERGE, a unified model for image generation and depth estimation, starting from a fixed pre-trained text-to-image model.
  arXiv  Detail & Related papers  (2025-10-27T17:44:56Z)
• TTS-VAR: A Test-Time Scaling Framework for Visual Auto-Regressive Generation [48.61855865678161]
  We present the first general test-time scaling framework for visual auto-regressive ( VAR) models.<n>We propose clustering-based diversity search and resampling-based potential selection.<n>Experiments on the powerful VAR model Infinity show a notable 8.7% GenEval score improvement.
  arXiv  Detail & Related papers  (2025-07-24T16:04:55Z)
• Spatial Reasoning with Denoising Models [49.83744014336816]
  We introduce a framework to perform reasoning over sets of continuous variables via denoising generative models.<n>For the first time, that order of generation can successfully be predicted by the denoising network itself.<n>Using these findings, we can increase the accuracy of specific reasoning tasks from 1% to >50%.
  arXiv  Detail & Related papers  (2025-02-28T14:08:30Z)
• Ambient Denoising Diffusion Generative Adversarial Networks for Establishing Stochastic Object Models from Noisy Image Data [4.069144210024564]
  We propose an augmented DDGAN architecture, Ambient DDGAN (ADDGAN) for learning realistic SOMs from noisy image data.<n>The ability of the proposed ADDGAN to learn realistic SOMs from noisy image data is demonstrated.
  arXiv  Detail & Related papers  (2025-01-31T12:40:43Z)
• Optimizing Resource Consumption in Diffusion Models through Hallucination Early Detection [87.22082662250999]
  We introduce HEaD (Hallucination Early Detection), a new paradigm designed to swiftly detect incorrect generations at the beginning of the diffusion process. We demonstrate that using HEaD saves computational resources and accelerates the generation process to get a complete image. Our findings reveal that HEaD can save up to 12% of the generation time on a two objects scenario.
  arXiv  Detail & Related papers  (2024-09-16T18:00:00Z)
• Assessing Sample Quality via the Latent Space of Generative Models [44.59115390303591]
  We propose to examine the latent space of a trained generative model to infer generated sample quality. This is feasible because the quality a generated sample directly relates to the amount of training data resembling it. We show that the proposed score correlates highly with the sample quality for various generative models including VAEs, GANs and Latent Diffusion Models.
  arXiv  Detail & Related papers  (2024-07-21T14:05:06Z)
• One-Line-of-Code Data Mollification Improves Optimization of Likelihood-based Generative Models [17.47235124122244]
  Likelihood-based GMs are attractive due to the possibility to generate new data by a single model evaluation. They typically achieve lower sample quality compared to state-of-the-art score-based diffusion models (DMs)
  arXiv  Detail & Related papers  (2023-05-30T09:58:47Z)
• Robust Compressed Sensing MRI with Deep Generative Priors [84.69062247243953]
  We present the first successful application of the CSGM framework on clinical MRI data. We train a generative prior on brain scans from the fastMRI dataset, and show that posterior sampling via Langevin dynamics achieves high quality reconstructions.
  arXiv  Detail & Related papers  (2021-08-03T08:52:06Z)

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.