PulseAugur
EN
LIVE 13:19:26

New research advances generative models for efficiency and evaluation

Several recent research papers explore advancements in generative models, focusing on improving their efficiency, evaluability, and alignment. One paper proposes a new framework for weighted sampling using score-based generative models, achieving significant speedups. Another theoretical framework addresses the statistical evaluability of generative models, distinguishing between metrics that can be reliably estimated from finite samples and those that cannot. Other research introduces methods for parameter-efficient generative modeling, calibrating models to distributional constraints, and aligning few-step generative models using sample-based variational inference. AI

IMPACT These papers introduce novel theoretical frameworks and practical methods for improving generative models, potentially leading to more efficient and reliable AI applications.

RANK_REASON The cluster consists of multiple academic papers published on arXiv, detailing theoretical frameworks and new methods for generative models.

Read on arXiv cs.AI →

AI-generated summary · Google Gemini · from 10 sources. How we write summaries →

COVERAGE [10]

  1. arXiv cs.AI TIER_1 English(EN) · Jiaming Song, Linqi Zhou ·

    Ideas in Inference-time Scaling can Benefit Generative Pre-training Algorithms

    arXiv:2503.07154v3 Announce Type: replace-cross Abstract: Generative pre-training is often framed through a false dichotomy between autoregressive models for discrete signals and diffusion models for continuous signals. We argue that the dichotomy is false because it conflates mo…

  2. arXiv cs.LG TIER_1 English(EN) · Shashaank Aiyer, Yishay Mansour, Shay Moran, Han Shao ·

    A Theoretical Framework for Statistical Evaluability of Generative Models

    arXiv:2604.05324v2 Announce Type: replace Abstract: Statistical evaluation aims to estimate the generalization performance of a model using held-out i.i.d. test data sampled from the ground-truth distribution. In supervised learning settings such as classification, performance me…

  3. arXiv cs.AI TIER_1 English(EN) · Heasung Kim, Taekyun Lee, Hyeji Kim, Gustavo de Veciana ·

    Efficient Weighted Sampling via Score-based Generative Models

    arXiv:2502.04646v2 Announce Type: replace-cross Abstract: Weighted sampling -- sampling from a probability density function (PDF) proportional to the product of a base PDF and a weight function -- is a fundamental technique with wide-ranging applications in variance reduction, bi…

  4. arXiv cs.LG TIER_1 English(EN) · Sascha Diefenbacher, Sofia Palacios Schweitzer, Gregor Kasieczka ·

    Generative Models and Statistical Validation

    arXiv:2605.30453v1 Announce Type: cross Abstract: Generative machine learning has become an essential tool in theoretical and experimental physics, especially in the context of fast surrogates and density estimators. In this work, we first introduce the underlying framework of mo…

  5. arXiv cs.LG TIER_1 English(EN) · Daniil Shlenskii, Nikita Gushchin, Lev Novitskiy, Dmitry V. Dylov, Alexander Korotin ·

    Midpoint Generative Models

    arXiv:2605.29920v1 Announce Type: new Abstract: We introduce Midpoint Generative Models (MGM), a principled framework for training one-step generative models. MGM is based on a simple symmetry of Flow Matching with linear interpolation: when the two endpoint distributions coincid…

  6. Hugging Face Daily Papers TIER_1 English(EN) ·

    Parameter-Efficient Generative Modeling with Controlled Vector Fields

    We introduce a continuous-time generative modeling framework, motivated by the Chow-Rashevskii theorem, that builds expressive flows from a small set of fixed vector fields and learned scalar controls. Instead of learning an unconstrained high-dimensional vector field, our framew…

  7. arXiv cs.AI TIER_1 English(EN) · Jaewoo Lee, Hyeongyu Kang, Dohyun Kim, Kyuil Sim, Woocheol Shin, Minsu Kim, Taeyoung Yun, Jeongjae Lee, Sanghyeok Choi, Tabitha Edith Lee, Jongchul Ye, Jinkyoo Park ·

    Aligning Few-Step Generative Models by Amortizing Sample-based Variational Inference

    arXiv:2605.26552v1 Announce Type: cross Abstract: Aligning a few-step generative model is challenging, since existing alignment frameworks typically rely on restrictive assumptions: a tractable likelihood, a specific ODE/SDE solver, or a particular model family. We introduce FAV,…

  8. arXiv stat.ML TIER_1 English(EN) · Jiguang Li, Sid Kankanala, Veronika Rockova ·

    Empirical Likelihood with Generative AI

    arXiv:2606.00425v1 Announce Type: cross Abstract: Moment conditions are widely used to identify parameters in models where the full likelihood is either unknown or intentionally left unspecified. Empirical likelihood methods address this problem by assigning probability weights t…

  9. arXiv stat.ML TIER_1 English(EN) · Veronika Rockova ·

    Empirical Likelihood with Generative AI

    Moment conditions are widely used to identify parameters in models where the full likelihood is either unknown or intentionally left unspecified. Empirical likelihood methods address this problem by assigning probability weights to the observed data so that the sample moment cond…

  10. arXiv stat.ML TIER_1 English(EN) · Henry D. Smith, Nathaniel L. Diamant, Brian L. Trippe ·

    Calibrating Generative Models to Distributional Constraints

    arXiv:2510.10020v4 Announce Type: replace Abstract: Generative models frequently suffer miscalibration, wherein statistics of the sampling distribution, such as the fraction of generations in a given class, deviate from desired values. We frame calibration as a constrained optimi…