Assessing Sample Quality in Conditional Generation under Compositional Shift

• URL: http://arxiv.org/abs/2606.09601v2
• Date: Tue, 09 Jun 2026 16:00:11 GMT
• Title: Assessing Sample Quality in Conditional Generation under Compositional Shift
• Abstract summary: We develop a per-sample trust score for assessing conditional samples using only the training distribution.<n>We show that the score can recover meaningful comparisons across extrapolated generations.<n>In biological imaging, selected samples preserve real morphological structure better and improve downstream predictive performance.
• Score: 34.24064416169904
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Conditional generators provide a natural tool for controllable generation, including settings where the desired condition is a new composition of observed attributes or experimental factors. In many applications, especially in scientific domains, such models are attractive to explore conditions for which real samples are rare, expensive, or not yet observed. However, this creates a circularity for evaluation: standard conditional quality metrics require a reference target distribution, but in the extrapolative regime that distribution is unavailable by definition. We address this problem with a post-hoc, per-sample trust score for assessing conditional samples using only the training distribution. The score combines two estimable quantities: global realism, measuring compatibility with the real data manifold, and attribute-wise faithfulness, measuring whether a sample is closer to the requested attributes than to plausible alternatives. We show that the score can recover meaningful comparisons across extrapolated generations, under a mild coverage condition on the observed attributes. These comparisons enable effective filtering, ranking, and abstention of generations and can be used directly on off-the-shelf pretrained models. In biological imaging, selected samples preserve real morphological structure better and improve downstream predictive performance, while similar gains are observed on controlled vision benchmarks. Finally, we show how the score can be applied during generation, enabling abstention before full decoding. Code is available at https://github.com/berkerdemirel/faithful-cond-gen.

Related papers

• Ramen: Robust Test-Time Adaptation of Vision-Language Models with Active Sample Selection [45.20212930761406]
  We present Ramen, a framework for robust test-time adaptation through active sample selection.<n>For each incoming test sample, Ramen retrieves a customized batch of relevant samples from previously seen data.<n>Experiments on multiple image corruption and domain-shift benchmarks demonstrate that Ramen achieves strong and consistent performance.
  arXiv  Detail & Related papers  (2026-04-23T14:33:27Z)
• Theory on Score-Mismatched Diffusion Models and Zero-Shot Conditional Samplers [49.97755400231656]
  We present the first performance guarantee with explicit dimensional dependencies for general score-mismatched diffusion samplers.<n>We show that score mismatches result in an distributional bias between the target and sampling distributions, proportional to the accumulated mismatch between the target and training distributions.<n>This result can be directly applied to zero-shot conditional samplers for any conditional model, irrespective of measurement noise.
  arXiv  Detail & Related papers  (2024-10-17T16:42:12Z)
• Assessing Sample Quality via the Latent Space of Generative Models [44.59115390303591]
  We propose to examine the latent space of a trained generative model to infer generated sample quality. This is feasible because the quality a generated sample directly relates to the amount of training data resembling it. We show that the proposed score correlates highly with the sample quality for various generative models including VAEs, GANs and Latent Diffusion Models.
  arXiv  Detail & Related papers  (2024-07-21T14:05:06Z)
• Self-Guided Generation of Minority Samples Using Diffusion Models [57.319845580050924]
  We present a novel approach for generating minority samples that live on low-density regions of a data manifold. Our framework is built upon diffusion models, leveraging the principle of guided sampling. Experiments on benchmark real datasets demonstrate that our approach can greatly improve the capability of creating realistic low-likelihood minority instances.
  arXiv  Detail & Related papers  (2024-07-16T10:03:29Z)
• Zero-Shot Conditioning of Score-Based Diffusion Models by Neuro-Symbolic Constraints [1.1826485120701153]
  We propose a method that, given a pre-trained unconditional score-based generative model, samples from the conditional distribution under arbitrary logical constraints.<n>We show how to manipulate the learned score in order to sample from an un-normalized distribution conditional on a user-defined constraint.<n>We define a flexible and numerically stable neuro-symbolic framework for encoding soft logical constraints.
  arXiv  Detail & Related papers  (2023-08-31T08:25:47Z)
• Variational Classification [51.2541371924591]
  We derive a variational objective to train the model, analogous to the evidence lower bound (ELBO) used to train variational auto-encoders. Treating inputs to the softmax layer as samples of a latent variable, our abstracted perspective reveals a potential inconsistency. We induce a chosen latent distribution, instead of the implicit assumption found in a standard softmax layer.
  arXiv  Detail & Related papers  (2023-05-17T17:47:19Z)
• Breaking the Spurious Causality of Conditional Generation via Fairness Intervention with Corrective Sampling [77.15766509677348]
  Conditional generative models often inherit spurious correlations from the training dataset. This can result in label-conditional distributions that are imbalanced with respect to another latent attribute. We propose a general two-step strategy to mitigate this issue.
  arXiv  Detail & Related papers  (2022-12-05T08:09:33Z)
• Fake It Till You Make It: Near-Distribution Novelty Detection by Score-Based Generative Models [54.182955830194445]
  existing models either fail or face a dramatic drop under the so-called near-distribution" setting. We propose to exploit a score-based generative model to produce synthetic near-distribution anomalous data. Our method improves the near-distribution novelty detection by 6% and passes the state-of-the-art by 1% to 5% across nine novelty detection benchmarks.
  arXiv  Detail & Related papers  (2022-05-28T02:02:53Z)
• On conditional versus marginal bias in multi-armed bandits [105.07190334523304]
  The bias of the sample means of the arms in multi-armed bandits is an important issue in adaptive data analysis. We characterize the sign of the conditional bias of monotone functions of the rewards, including the sample mean. Our results hold for arbitrary conditioning events and leverage natural monotonicity properties of the data collection policy.
  arXiv  Detail & Related papers  (2020-02-19T20:16:10Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.