Pith. sign in

REVIEW 12 cited by

Improved Precision and Recall Metric for Assessing Generative Models

Not yet reviewed by Pith; the record is open.

Re-run · record.json Download PDF Read on arXiv ↗

This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.

SPECIMEN: schema-true, not a live event

T0 review · schema-true

One-sentence machine reading of the paper's core claim.

pith:XXXXXXXX · record.json · timestamp

arxiv 1904.06991 v3 pith:2NLZQLG5 submitted 2019-04-15 stat.ML cs.LGcs.NE

Tuomas Kynkäänniemi , Tero Karras , Samuli Laine , Jaakko Lehtinen , Timo Aila This is my paper
classification stat.MLcs.LGcs.NE
keywords metricestimategenerativeidentifyimprovedmethodsmodelquality
verification ladder T0 review T1 audit T2 compute T3 formal
0 comments
read the original abstract

The ability to automatically estimate the quality and coverage of the samples produced by a generative model is a vital requirement for driving algorithm research. We present an evaluation metric that can separately and reliably measure both of these aspects in image generation tasks by forming explicit, non-parametric representations of the manifolds of real and generated data. We demonstrate the effectiveness of our metric in StyleGAN and BigGAN by providing several illustrative examples where existing metrics yield uninformative or contradictory results. Furthermore, we analyze multiple design variants of StyleGAN to better understand the relationships between the model architecture, training methods, and the properties of the resulting sample distribution. In the process, we identify new variants that improve the state-of-the-art. We also perform the first principled analysis of truncation methods and identify an improved method. Finally, we extend our metric to estimate the perceptual quality of individual samples, and use this to study latent space interpolations.

Discussion (0). Sign in to comment.

Forward citations

Cited by 12 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. High-Resolution Image Synthesis with Latent Diffusion Models

    cs.CV 2021-12 conditional novelty 7.0 of 10

    Latent diffusion models achieve state-of-the-art inpainting and competitive results on unconditional generation, scene synthesis, and super-resolution by performing the diffusion process in the latent space of pretrai...

  2. GLIDE: Towards Photorealistic Image Generation and Editing with Text-Guided Diffusion Models

    cs.CV 2021-12 accept novelty 7.0 of 10

    A 3.5-billion-parameter diffusion model with classifier-free guidance generates images preferred over DALL-E by human raters and can be fine-tuned for text-guided inpainting.

  3. Diffusion Models Beat GANs on Image Synthesis

    cs.LG 2021-05 accept novelty 7.0 of 10

    Diffusion models with architecture improvements and classifier guidance achieve superior FID scores to GANs on unconditional and conditional ImageNet image synthesis.

  4. FairDiffuseVQVAE: Sampling-Time Fairness in Tabular Diffusion via Conditional Refinement of Vector-Quantized Latents

    cs.LG 2026-07 conditional novelty 6.0 of 10

    FairDiffuseVQVAE reaches state-of-the-art fairness on the standard tabular benchmark (DPR 0.702, EOR 0.686) by uniform protected-attribute sampling at inference, paying ~15 AUC points of utility.

  5. GenSyn10: A Multi-Generative AI Dataset For Benchmarking Image Classification

    cs.CV 2026-07 conditional novelty 6.0 of 10

    GenSyn10 provides 60k CIFAR-10-aligned images from FLUX.2, HunyuanImage-3.0, and Qwen-Image-2512, showing detectors lose 4–18 points of accuracy on an unseen generator.

  6. VideoMDM: Towards 3D Human Motion Generation From 2D Supervision

    cs.LG 2026-06 unverdicted novelty 6.0 of 10

    VideoMDM learns coherent 3D motion manifolds from 2D supervision alone by using a pretrained lifter as noisy teacher, depth-weighted 2D reprojection loss, and adapted regularizers, nearly matching fully 3D-supervised ...

  7. IDEAL: In-DEpth ALignment Makes A Discrete Representation AutoEncoder

    cs.CV 2026-06 unverdicted novelty 6.0 of 10

    IDEAL improves discrete representation autoencoders by jointly aligning quantized tokens with shallow and deep VFM features, reporting 0.61 rFID on ImageNet and 1.89 gFID for autoregressive image generation.

  8. Score-Control for Hallucination Reduction in Diffusion Models

    cs.CV 2026-05 unverdicted novelty 6.0 of 10

    VSM modulates the score Jacobian using variance guidance to reduce hallucinations in diffusion models by up to 25% on synthetic and real datasets while preserving fidelity and diversity.

  9. Phantom Evidence: How and Why Generative AI Manufactures False Positives in Science

    q-bio.NC 2026-07 conditional novelty 5.0 of 10

    Evidence from a convincing-looking AI output is bounded by how often such outputs occur when the target is absent; generative AI inflates that denominator, and the paper calls the resulting overestimate 'phantom evidence'.

  10. HistoFID- Calibrating Frechet-distance evaluation across pathology foundation models

    eess.IV 2026-07 conditional novelty 5.0 of 10

    Raw Fréchet distances in pathology vary ~30-fold across encoders; dividing by each encoder's own within-cohort floor cuts cross-encoder variation by ~89% within and ~58% across cohorts.

  11. GRASP: Guided Residual Adapters with Sample-wise Partitioning

    cs.CV 2025-12 unverdicted novelty 5.0 of 10

    GRASP applies deterministic conditioning-space partitioning and sample-wise residual adapters to improve tail-class fidelity, diversity, and downstream utility in flow matching models, outperforming full fine-tuning a...

  12. Sealing The Backdoor: Unlearning Adversarial Text Triggers In Diffusion Models Using Knowledge Distillation

    cs.CV 2025-08 unverdicted novelty 5.0 of 10

    SKD-CAG erases adversarial text triggers from diffusion models by distilling the model's own clean outputs through cross-attention guidance, claiming 100% and 93% removal for pixel and style backdoors.

Pith tools