mlsanityData to 7 Oct 2026

Divergence Frontiers for Generative Models: Sample Complexity, Quantization Effects, and Frontier Integrals

arXiv 2106.07898 ↗First posted 15 Jun 2021stat.ML cs.LGcited by 18

Not on the map. Why?

Abstract (arXiv)

The spectacular success of deep generative models calls for quantitative tools to measure their statistical performance. Divergence frontiers have recently been proposed as an evaluation framework for generative models, due to their ability to measure the quality-diversity trade-off inherent to deep generative modeling. We establish non-asymptotic bounds on the sample complexity of divergence frontiers. We also introduce frontier integrals which provide summary statistics of divergence frontiers. We show how smoothed estimators such as Good-Turing or Krichevsky-Trofimov can overcome the missing mass problem and lead to faster rates of convergence. We illustrate the theoretical results with numerical examples from natural language processing and computer vision.

Key facts. Cited by 18 papers in the map (data through 7 Oct 2026). Builds on 23 of its 68 references that are in the map. First posted on arXiv 15 Jun 2021.

Builds on

23 of 68 references.

References · 23

  1. 2015Deep Residual Learning for Image Recognitioncited by 51,809
  2. 2020Language Models are Few-Shot Learnerscited by 32,199
  3. 2018A Style-Based Generator Architecture for Generative Adversarial Networkscited by 5,986
  4. 2016Pointer Sentinel Mixture Modelscited by 2,992
  5. 2016Improved Techniques for Training GANscited by 2,938
  6. 2018Glow: Generative Flow with Invertible 1x1 Convolutionscited by 1,539
  7. 2019Generating Diverse High-Fidelity Images with VQ-VAE-2cited by 1,376
  8. 2018Demystifying MMD GANscited by 1,268
  9. 2020Training Generative Adversarial Networks with Limited Datacited by 1,184
  10. 2016Least Squares Generative Adversarial Networkscited by 1,000
  11. 2019Improved Precision and Recall Metric for Assessing Generative Modelscited by 925
  12. 2019Defending Against Neural Fake Newscited by 553
11 more
  1. 2020Jukebox: A Generative Model for Musiccited by 535
  2. 2021MAUVE: Measuring the Gap Between Neural Text and Human Text using Divergence Frontierscited by 466
  3. 2020Reliable Fidelity and Diversity Metrics for Generative Modelscited by 425
  4. 2018Assessing Generative Models via Precision and Recallcited by 401
  5. 2016Revisiting Classifier Two-Sample Testscited by 258
  6. 2018Spreading vectors for similarity searchcited by 114
  7. 2019Practical and Consistent Estimation of f-Divergencescited by 24
  8. 2020Evaluation Metrics for Conditional Image Generationcited by 18
  9. 2021Non-Asymptotic Performance Guarantees for Neural Estimation of mathsff-Divergencescited by 15
  10. 2018Quantitatively Evaluating GANs With Divergences Proposed for Trainingcited by 4
  11. 2020Deep Residual Mixture Modelscited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2022: 3; 2023: 1; 2024: 6; 2025: 5; 2026: 3.

18 papers.

Sort:

By area:

  1. 2023Does Writing with Language Models Reduce Content Diversity?cited by 190
  2. 2025Remasking Discrete Diffusion Models with Inference-Time Scalingcited by 181
  3. 2024DistiLLM: Towards Streamlined Distillation for Large Language Modelscited by 145
  4. 2022MAUVE Scores for Generative Models: Theory and Practicecited by 34
  5. 2025Continuously Augmented Discrete Diffusion model for Categorical Generative Modelingcited by 32
  6. 2024A Practical Guide to Sample-based Statistical Distances for Evaluating Generative Models in Sciencecited by 14
  7. 2024SynthesizRR: Generating Diverse Datasets with Retrieval Augmentationcited by 13
  8. 2024KnowledgeSG: Privacy-Preserving Synthetic Text Generation with Knowledge Distillation from Servercited by 11
  9. 2025InvisibleInk: High-Utility and Low-Cost Text Generation with Differential Privacycited by 10
  10. 2025Enabling Approximate Joint Sampling in Diffusion LMscited by 9
  11. 2022RenyiCL: Contrastive Representation Learning with Skew Renyi Divergencecited by 8
  12. 2022On the Usefulness of Embeddings, Clusters and Strings for Text Generator Evaluationcited by 7
6 more
  1. 2024CorrSynth -- A Correlated Sampling Method for Diverse Dataset Generation from LLMscited by 3
  2. 2024Style-News: Incorporating Stylized News Generation and Adversarial Verification for Neural Fake News Detectioncited by 2
  3. 2026Fixed-Point Masked Generative Modelingcited by 2
  4. 2026Privately Fine-Tuned LLMs Preserve Temporal Dynamics in Tabular Datacited by 1
  5. 2026Gumbel Distillation for Parallel Text Generationcited by 1
  6. 2025Token Homogenization under Positional Biascited by 0