mlsanityData to 7 Oct 2026

DAK-UCB: Diversity-Aware Prompt Routing for LLMs and Generative Models

arXiv 2603.23140 ↗First posted 24 Mar 2026cs.LGcited by 2

Not on the map. Why?

Abstract (arXiv)

The expansion of generative AI and LLM services underscores the growing need for adaptive mechanisms to select an appropriate available model to respond to a user's prompts. Recent works have proposed offline and online learning formulations to identify the optimal generative AI model for an input prompt, based solely on maximizing prompt-based fidelity evaluation scores, e.g., CLIP-Score in text-to-image generation. However, such fidelity-based selection methods overlook the diversity of generated outputs, and hence, they can fail to address potential diversity shortcomings in the generated responses. In this paper, we introduce the Diversity-Aware Kernelized Upper Confidence Bound (DAK-UCB) method as a contextual bandit algorithm for the online selection of generative models with diversity considerations. The proposed DAK-UCB method incorporates both fidelity and diversity-related metrics into the selection process. We design this framework based on prompt-aware diversity score functions that decompose to a two-sample-based expectation over prompt-output pairs in the previous generation rounds. Specifically, we illustrate the application of our framework using joint kernel distance and kernel entropy measures. Our experimental results demonstrate the effectiveness of DAK-UCB in promoting diversity-aware model selection while maintaining fidelity in the generations for a sequence of prompts. The code is available at https://github.com/Donya-Jafari/DAK-UCB.

Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 40 of its 59 references that are in the map. First posted on arXiv 24 Mar 2026.

Builds on

40 of 59 references.

References · 40

  1. 2021Learning Transferable Visual Models From Natural Language Supervisioncited by 33,456
  2. 2014Microsoft COCO: Common Objects in Contextcited by 17,498
  3. 2023DINOv2: Learning Robust Visual Features without Supervisioncited by 7,764
  4. 2023BLIP-2: Bootstrapping Language-Image Pre-training with Frozen Image Encoders and Large Language Modelscited by 6,069
  5. 2018Demystifying MMD GANscited by 1,268
  6. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  7. 2023FrugalGPT: How to Use Large Language Models While Reducing Cost and Improving Performancecited by 791
  8. 2023Scaling up GANs for Text-to-Image Synthesiscited by 492
  9. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 425
  10. 2018Assessing Generative Models via Precision and Recallcited by 401
  11. 2022The Vendi Score: A Diversity Evaluation Metric for Machine Learningcited by 299
  12. 2023Exposing flaws of generative model evaluation metrics and their unfair treatment of diffusion modelscited by 199
28 more
  1. 2013Finite-Time Analysis of Kernelised Contextual Banditscited by 177
  2. 2023CADS: Unleashing the Diversity of Diffusion Models through Condition-Annealed Samplingcited by 124
  3. 2018Practical Contextual Bandits with Regression Oraclescited by 92
  4. 2022Generating High Fidelity Data from Low-density Regions using Diffusion Modelscited by 78
  5. 2019Randomized Exploration in Generalized Linear Banditscited by 70
  6. 2023Particle Guidance: non-I.I.D. Diverse Sampling with Diffusion Modelscited by 65
  7. 2020Thompson Sampling Algorithms for Mean-Variance Banditscited by 27
  8. 2025Prompt-to-Leaderboardcited by 27
  9. 2024Towards a Scalable Reference-Free Evaluation of Generative Modelscited by 25
  10. 2024Kandinsky 3: Text-to-Image Synthesis for Multifunctional Generative Frameworkcited by 25
  11. 2024Improving Geo-diversity of Generated Images with Contextualized Vendi Score Guidancecited by 24
  12. 2024An Interpretable Evaluation of Entropy-based Novelty of Generative Modelscited by 17
  13. 2024Unveiling Differences in Generative Models: A Scalable Differential Clustering Approachcited by 16
  14. 2024Scendi Score: Prompt-Aware Diversity Evaluation via Schur Complement of CLIP Embeddingscited by 15
  15. 2020Near-linear Time Gaussian Process Optimization with Adaptive Batching and Resparsificationcited by 13
  16. 2024Be More Diverse than the Most Diverse: Optimal Mixtures of Generative Models via Mixture-UCB Bandit Algorithmscited by 13
  17. 2024ProCreate, Don't Reproduce! Propulsive Energy Diffusion for Creative Generationcited by 12
  18. 2024Do Vendi Scores Converge with Finite Samples? Truncated Vendi Score for Finite-Sample Convergence Guaranteescited by 12
  19. 2025Kernel-based Unsupervised Embedding Alignment for Enhanced Visual Representation in Vision-language Modelscited by 12
  20. 2024PAK-UCB Contextual Bandit: An Online Learning Approach to Prompt-Aware Selection of Generative Models and LLMscited by 11
  21. 2025Towards an Explainable Comparison and Alignment of Feature Embeddingscited by 9
  22. 2024Almost Minimax Optimal Best Arm Identification in Piecewise Stationary Linear Banditscited by 6
  23. 2021Johnson-Lindenstrauss Embeddings with Kronecker Structurecited by 5
  24. 2022Multi-Armed Bandits with Self-Information Rewardscited by 5
  25. 2024Best Arm Identification with Minimal Regretcited by 5
  26. 2025Optimal Multi-Objective Best Arm Identification with Fixed Confidencecited by 4
  27. 2025PromptWise: Online Learning for Cost-Aware Prompt Assignment in Generative Modelscited by 3
  28. 2026MMD Guidance: Training-Free Distribution Adaptation for Diffusion Models via Maximum Mean Discrepancy Guidancecited by 3

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 2.

2 papers.

  1. 2026KODA: Contrastive Representation Comparison and Alignment for Vision-Language Foundation Modelscited by 2
  2. 2026Exploring More to Solve More: Boosting Diversity in Text Diffusion Models via Entropy-Based Guidancecited by 1