English
Related papers

Related papers: Conditional Vendi Score: An Information-Theoretic …

200 papers

Diversity is an important criterion for many areas of machine learning (ML), including generative modeling and dataset curation. However, existing metrics for measuring diversity are often domain-specific and limited in flexibility. In this…

Machine Learning · Computer Science 2023-07-04 Dan Friedman , Adji Bousso Dieng

Diffusion models have demonstrated remarkable success in high-fidelity image synthesis and prompt-guided generative modeling. However, ensuring adequate diversity in generated samples of prompt-guided diffusion models remains a challenge,…

Computer Vision and Pattern Recognition · Computer Science 2025-10-31 Mohammad Jalali , Haoyu Lei , Amin Gohari , Farzan Farnia

Deep generative models have achieved great success in producing high-quality samples, making them a central tool across machine learning applications. Beyond sample quality, an important yet less systematically studied question is whether…

Machine Learning · Computer Science 2026-02-17 Farzan Farnia , Mohammad Jalali , Azim Ospanov

Quality and diversity have been proposed as reasonable heuristics for assessing content generated by co-creative systems, but to date there has been little agreement around what constitutes the latter or how to measure it. Proposed…

Computer Vision and Pattern Recognition · Computer Science 2024-03-22 Francisco Ibarrola , Kazjon Grace

Despite advances in generation quality, current text-to-image (T2I) models often lack diversity, generating homogeneous outputs. This work introduces a framework to address the need for robust diversity evaluation in T2I models. Our…

Computer Vision and Pattern Recognition · Computer Science 2025-11-14 Isabela Albuquerque , Ira Ktena , Olivia Wiles , Ivana Kajić , Amal Rannen-Triki , Cristina Vasconcelos , Aida Nematzadeh

Synthesis of digital artifacts conditioned on user prompts has become an important paradigm facilitating an explosion of use cases with generative AI. However, such models often fail to connect the generated outputs and desired target…

Machine Learning · Computer Science 2026-04-15 Melvin Wong , Yew-Soon Ong , Abhishek Gupta , Kavitesh K. Bali , Caishun Chen

The limits of open-ended generative models are unclear, yet increasingly important. What causes them to succeed and what causes them to fail? In this paper, we take a prompt-centric approach to analyzing and bounding the abilities of…

Computation and Language · Computer Science 2023-02-21 Albert Lu , Hongxin Zhang , Yanzhe Zhang , Xuezhi Wang , Diyi Yang

Prompt-guided generative AI models have rapidly expanded across vision and language domains, producing realistic and diverse outputs from textual inputs. The growing variety of such models, trained with different data and architectures,…

Machine Learning · Computer Science 2026-02-09 Mehdi Lotfian , Mohammad Jalali , Farzan Farnia

Error prediction in large language models often relies on domain-specific information. In this paper, we present measures for quantification of error in the response of a large language model based on the diversity of responses to a given…

Computation and Language · Computer Science 2024-01-30 Noel Ngu , Nathaniel Lee , Paulo Shakarian

Recent advances in text-to-image (T2I) generation have achieved remarkable visual outcomes through large-scale rectified flow models. However, how these models behave under long prompts remains underexplored. Long prompts encode rich…

Computer Vision and Pattern Recognition · Computer Science 2025-11-26 Bo-Kai Ruan , Teng-Fang Hsiao , Ling Lo , Yi-Lun Wu , Hong-Han Shuai

Following the initial excitement, Text-to-Image (TTI) models are now being examined more critically. While much of the discourse has focused on biases and stereotypes embedded in large-scale training datasets, the sociotechnical dynamics of…

Human-Computer Interaction · Computer Science 2025-04-22 Maria-Teresa De Rosa Palmini , Eva Cetinic

Evaluating the diversity of generative models without reference data poses methodological challenges. The reference-free Vendi and RKE scores address this by quantifying the diversity of generated data using matrix-based entropy measures.…

Machine Learning · Statistics 2025-06-25 Azim Ospanov , Farzan Farnia

The purpose of unconditional text generation is to train a model with real sentences, then generate novel sentences of the same quality and diversity as the training data. However, when different metrics are used for comparing the methods…

Computation and Language · Computer Science 2020-07-03 Ping Cai , Xingyuan Chen , Peng Jin , Hongjun Wang , Tianrui Li

Determining conditional independence (CI) relationships between random variables is a fundamental yet challenging task in machine learning and statistics, especially in high-dimensional settings. Existing generative model-based CI testing…

Machine Learning · Computer Science 2025-05-30 Yixin Ren , Chenghou Jin , Yewei Xia , Li Ke , Longtao Huang , Hui Xue , Hao Zhang , Jihong Guan , Shuigeng Zhou

Deep generative modeling of natural languages has achieved many successes, such as producing fluent sentences and translating from one language into another. However, the development of generative modeling techniques for paraphrase…

Computation and Language · Computer Science 2023-11-28 Haotian Luo , Yixin Liu , Peidong Liu , Xianggen Liu

We introduce GRADE, an automatic method for quantifying sample diversity in text-to-image models. Our method leverages the world knowledge embedded in large language models and visual question-answering systems to identify relevant…

Computer Vision and Pattern Recognition · Computer Science 2025-03-12 Royi Rassin , Aviv Slobodkin , Shauli Ravfogel , Yanai Elazar , Yoav Goldberg

This work presents an open-source unified benchmarking and evaluation framework for text-to-image generation models, with a particular focus on the impact of metadata augmented prompts. Leveraging the DeepFashion-MultiModal dataset, we…

Graphics · Computer Science 2025-05-09 Kapil Wanaskar , Gaytri Jena , Magdalini Eirinaki

Prompts used in recent large language model based applications are often fixed and lengthy, leading to significant computational overhead. To address this challenge, we propose Generative Prompt Internalization (GenPI), a lightweight method…

Computation and Language · Computer Science 2026-02-12 Haebin Shin , Lei Ji , Yeyun Gong , Sungdong Kim , Eunbi Choi , Minjoon Seo

Large language models (LLMs) have been recently leveraged as training data generators for various natural language processing (NLP) tasks. While previous research has explored different approaches to training models using generated data,…

Computation and Language · Computer Science 2023-10-19 Yue Yu , Yuchen Zhuang , Jieyu Zhang , Yu Meng , Alexander Ratner , Ranjay Krishna , Jiaming Shen , Chao Zhang

Effective generalization in language models depends critically on the diversity of their training data. Yet existing diversity metrics often fall short of this goal, relying on surface-level heuristics that are decoupled from model…

‹ Prev 1 2 3 10 Next ›