Nov 2025· GECCO Companion· 0 citations· 48 references
Computer Science
TL;DR
The application of inference-time prompt-embedding optimization for text-to-image generation using Stable Diffusion XL Turbo indicates that evolutionary optimization is a robust method for exploring embedding spaces without the need for gradients or model fine-tuning.
Abstract
Deep generative models have progressively improved image generation, particularly in terms of realism and semantic alignment. Despite this, there is still room for improvement, as illustrated by strategies such as prompt engineering. In this context, we studied the application of inference-time prompt-embedding optimization for text-to-image generation using Stable Diffusion XL Turbo. We compared the gradient-based Adam optimizer with the evolutionary sep-CMA-ES using a weighted objective that combined the LAION Aesthetic Predictor V2 and CLIPScore. sep-CMA-ES consistently achieved higher objective scores than Adam while requiring fewer computational resources. The findings indicate that evolutionary optimization is a robust method for exploring embedding spaces without the need for gradients or model fine-tuning.
GAOKAO-Bench is introduced, an intuitive benchmark that employs questions from the Chinese GAOKAO examination as test samples, including both subjective and objective questions that contribute a robust evaluation benchmark for future large language models and offers valuable insights into the advantages and limitations...
Xiaotian Zhang, Chun-yan Li, Yi Zong et al.· arXiv.org· 216 citations· ⚡17
An algorithm is proposed which leverages disentangled semantic factors to generate adversarial perturbation by altering controlled semantic attributes to fool the learner towards various "adversarial" targets.
Haonan Qiu, Chaowei Xiao, Lei Yang et al.· arXiv.org· 208 citations· ⚡17
This work proposes a HALO that directly maximizes the utility of generations instead of maximizing the log-likelihood of preferences, and suggests that the best loss depends on the inductive biases most appropriate for a given setting, an oft-overlooked consideration.
This work investigates the possibilities of using LLMs in a resume screening setting via a document retrieval framework that simulates job candidate selection and finds that the MTEs are biased, significantly favoring White-associated names in 85% of cases and female-associated names in only 11.1% of cases.
This work shows that orders of magnitude enhancement in performance could be obtained by a combination of hardware improvements and tight quantum-HPC integration and introduces high-performance architectures for quantum-probabilistic computing with custom-designed accelerators to tackle today's industry-scale classical...
Masoud Mohseni, Artur Scherer, K. Johnson et al.· arXiv.org· 121 citations· ⚡9
This paper presents a comprehensive overview of the Ultralytics YOLO family, emphasizing architectural evolution, benchmarking, deployment, and emerging directions from YOLOv5 through YOLO27, and examines detection, segmentation, depth, classification, pose, oriented detection, tracking, export, quantization, and deplo...
