Drifting Models

Follow this topic Ask about this topic

In generative modeling, drifting models learn a vector field—often using kernel-based attraction and repulsion—to transport a source distribution toward data in one or a few steps. Methods connect these fields to score matching and Wasserstein gradient flows, with variants based on Sinkhorn divergences, learned representations, symmetry, multiple bandwidths, and second-order dynamics. Theory examines identifiability, stability, convergence, and finite-sample bias; applications include image and molecule generation, speech enhancement, and robot policies. A related line studies models whose data or behavior changes over time, using drift detection, retraining, parameter extrapolation, and risk-aware monitoring, with open questions about reliable guarantees and when adaptation justifies its cost.

Latest papers

53 papers2 papers in the last four weeks, down 67% on the four weeks before. 0.0% of all new papers.

Multi-Bandwidth Distribution Matching Distillation: On the Equivalence of Distribution Matching Distillation and Drifting Models
Oct 7, 2026Jialin Zhu, Xing Liu, Feixiang He +1Distribution Matching Distillation Diffusion Model Distillation
Baidu Inc., China · Central South University, China · AI Centre, University College London, UK
Learning Discriminative Geometry for Drifting Models
Oct 3, 2026Doudou Zhang, Wenwen Hou, Yilin Chen +1Drifting Models Image Generation
ELLIS Institute Finland · Aalto University
Refresh or Realize? Compute Allocation in Drifting Models
Sep 26, 2026Sipeng Chen, Xu Zheng, Shibo LiDrifting Models
Department of Computer Science Florida State University · Knight Foundation School of Computing and Information Sciences Florida International University
One-Step Generative Surrogate Models via Block-Triangular Joint Drifting
Sep 22, 2026Nicholas Geissler, Shreya Jha, Ricardo Baptista +1Surrogate Modeling One-Step Generative Modeling
Courant Institute of Mathematical Sciences, New York University · Statistical Sciences, University of Toronto · Institute of Mathematics, EPFL
Concept Drift from a Causal Perspective
Sep 21, 2026Eduardo V. L. Barboza, Jean Paul Barddal, Robert Sabourin +1Concept Drift Drifting Models
LIVIA, École de Technologie Supérieure · Graduate Program in Informatics (PPGIa), Pontifícia Universidade Católica do Paraná (PUCPR)
Convergence rates for generative drifting flows: fixed-scale obstructions and multihead acceleration
Sep 14, 2026Arthur Stéphanovitch, Eddie AamariOne-Step Generative Modeling Generative Modeling
Centre de recherche en économie et statistique, ENSAE, IP Paris · Département de Mathématiques et Applications, CNRS, École Normale Supérieure, PSL
A Dominant Supplier Slows Recursive Drift More Than It Steers It
Sep 10, 2026Yangze Liu, Zhongyi HanDrifting Models Recursive Language Models
Shandong University
Evaluating Model Retraining under Drift: Paired Comparisons of Cumulative Subgroup Disparity
Sep 9, 2026Aaron CerossAlgorithmic Fairness Group Fairness
Birmingham Law School, University of Birmingham
When to Retrain: An Empirical Study of Retraining Policies for Streaming ML Under Concept Drift, Budget, and Latency Constraints
Aug 19, 2026Sawan DasariDrifting Models Concept Drift
Concept Drift Detection and Adaptive Retraining of Malware Classification Models
Aug 13, 2026Christofer Washington Berruz Chungata, Martin Jurecek, Katerina Potika +2Concept Drift Malware Classification
Department of Computer Science, San Jose State University · Faculty of Information Technology, Czech Technical University in Prague
Mapping and Measuring the Behavioral Evolution of Large Language Models
Aug 11, 2026Dong Qiao, Chris Ding, Jicong FanLLM Evaluation Representation Geometry in Language Models
School of Data Science, The Chinese University of Hong Kong, Shenzhen, China
Second Order Drifting Models
Aug 8, 2026Drake Brown, Yuhao Huang, Shih-Hsin Wang +1One-Step Generative Modeling Drifting Models
Department of Mathematics Scientific Computing and Imaging (SCI) Institute University of Utah
CFD-Guided Detection of Concept Drift in Multimodal Physiologic Signals
Aug 7, 2026Farouk Ganiyu Adewumi, Timothy Oladunni, Rochak Ghimire +3Concept Drift Electrocardiography
Computer Science, Morgan State University, Baltimore, USA · Computer Science, Fisk University, Nashville, USA · Electrical Engineering, Morgan State University, Baltimore, USA +1
NOMADD: Numerical Optimization of Models Adapting to Data Drift
Aug 3, 2026Swapn Shah, Keith BurghardtConcept Drift Tabular ML
University of North Carolina at Charlotte
Finite-Probe Total-Variation Certificates for Finite-Basis Drifting Models
Aug 3, 2026Sam Andersson, Ricky MolénUncertainty Quantification Drifting Models
Department of Clinical Neuroscience, Karolinska Institutet, Stockholm, Sweden · Department of Computational Science and Technology, KTH Royal Institute of Technology, Stockholm, Sweden
When Can You Correct Distribution Drift in Temporal Graph Generation? A Sharpening--Drift Tension and an Impossibility for Observation-Based Correction
Jul 27, 2026Tianpeng Li, Xuan Guo, Wenjun Wang +2Flow Matching Distribution Shift
College of New Media and Communication, Tianjin University · School of Artificial Intelligence, Tianjin University · School of Cyberspace Security,Hangzhou Dianzi University
Self-Supervised Consistency Enhanced Disentangled Learning for Neural Decoding Generalization in Brain-Machine Interface
Jul 27, 2026Jiyu Wei, Di Hong, Zhanjie Zhang +3Disentangled Representation Learning Brain-Computer Interfaces
College of Computer Science and Technology, Zhejiang University, Hangzhou, China · Nanhu Brain-Computer Interface Institute, Hangzhou, China
Steeringless Drifting: Differential-Torque Control of a Four-Wheel Independently Driven Vehicle
Jul 26, 2026Sheng Zhao, Zexin Wu, Dongyang Zhou +2Robotic Control Autonomous Driving
School of Mechanical Engineering, Shanghai Jiao Tong University, Shanghai 200240, China · School of Mechanical and Aerospace Engineering, Nanyang Technological University, Singapore 639798, Singapore
DriftWorld: Fast World Modeling through Drifting
Jul 16, 2026Susie Lu, Haonan Chen, Weirui Ye +1World Models for Robotics World Model-Based Planning
Massachusetts Institute of Technology · Harvard University
Certified World Models as Sensing Clocks: Drift-Aware Deadlines for Active Perception
Jul 1, 2026Hongbo WangActive Perception Active Sensing
Distilling Drifting Transformers with Representation Autoencoders
Jun 14, 2026Jiawei Zhang, Mengfei Xia, Gen Li +1Diffusion Model Distillation Autoencoders
Tsinghua University · Ant Group · CUHK
Who Drifted: the System or the Judge? Anytime-Valid Attribution in LLM Evaluation Pipelines
Jun 13, 2026Yitao LiLLM Evaluation LLM-as-a-Judge
DriftingMol: Decoder-Coupled Drift for One-Pass Property-Conditional Molecular Generation
May 24, 2026Jiangjie Qiu, Yijun Li, Wentao Li +1One-Step Generative Modeling Molecular Generation
Beijing Key Laboratory of Artificial Intelligence for Advanced Chemical Engineering Materials
Drift-Resistant Navigation World Model with Anchored Epipolar Guidance
May 23, 2026Po-Chien Luan, Zimin Xia, Wuyang Li +2World Models for Robotics Model-Based Planning
ContextEcho: A Benchmark for Persona Drift in Long Agentic-Coding Sessions
May 22, 2026Xianzhong Ding, Yangyang Yu, Changwei Liu +1Tool-Augmented Language Model Agents AI Coding Agents
Center for Advanced AI, Accenture
Finite-Particle Convergence Rates for Conservative and Non-Conservative Drifting Models
May 21, 2026Krishnakumar BalasubramanianOne-Step Generative Modeling Drifting Models
Department of Statistics, University of California, Davis
Drifting Objectives for Refining Discrete Diffusion Language Models
May 19, 2026Daisuke Oba, Hiroki Furuta, Naoaki OkazakiDiffusion Model Fine-Tuning Diffusion Language Models
Institute of Science Tokyo · AIST · NII LLMC
Counterfactual Explanations Under Concept Drift
May 17, 2026Marcin Kostrzewa, Jerzy Stefanowski, Maciej ZiębaCounterfactual Explanations Concept Drift
Wrocław University of Science and Technology · Pozna´n University of Technology · Tooploox
Drift Flow Matching
May 17, 2026Chenrui Ma, Xi Xiao, Lin Zhao +3Flow Matching Inference-Time Scaling
University of California, Irvine · University of Virginia · University of Alabama at Birmingham +1
DriftXpress: Faster Drifting Models via Projected RKHS Fields
May 12, 2026Ali Falahati, Elliot Creager, Gautam Kamath +1Image Generation One-Step Generative Modeling
Cheriton School of Computer Science, University of Waterloo and Vector Institute. · Department of Electrical and Computer Engineering, University of Waterloo and Vector Institute. · A*STAR Centre for Frontier AI Research, Singapore.

Oct 7, 2026J/K move · Enter open · S save

Multi-Bandwidth Distribution Matching Distillation: On the Equivalence of Distribution Matching Distillation and Drifting Models

Jialin Zhu, Xing Liu, Feixiang He, He Wang

Baidu Inc., China · Central South University, China · AI Centre, University College London, UK

Distribution Matching Distillation Diffusion Model Distillation One-Step Generative Modeling Drifting Models

Open paper AskPDF arXiv

Researchers are exploring effective one-step generative model continuously, and, Drifting Models (Deng et al., 2026), demonstrate great potential in one-step generation recently. There are works that reveal the connection between Diffusion & Flow Style Generative Models (DFSGMs) (Ho et al., 2020; Song et al., 2020a;b; Lipman et al., 2022; Liu et al., 2022) and Drifting Models (Li & Zhu, 2026; Lai et al., 2026; Turan et al., 2026). But no one has yet established a precise correspondence between the Drifting Model and the widely used distillation method- Distribution Matching Distillation (DMD/DMD2) (Yin et al., 2024b;a) to the best of our knowledge, even though their optimization objective formulas are virtually identical. In this paper, we prove that by converting the velocity-field / noise-field from the pre-trained DFSGMs into the attraction force field in Drifting Models and estimating the repulsion force field from the generative distribution, training the Drifting Model is naturally equivalent to the Distribution Matching Distillation. With this equivalent concept, we propose an improved method based on DMD from the Drifting Model's perspective- Multi-Bandwidth Distribution Matching Distillation (MBDMD).