Diffusion Inversion

Follow this topic Ask about this topic

Momentum

3 papers in the last four weeks, against 2 the four weeks before. 0.0% of all new papers.

Jul 6Week of Sep 21

Latest papers 24

All topics

Temporal Gradient Inversion for Private Trajectory Reconstruction in Embodied Reinforcement Learning
Sep 24, 2026Sudip Bhujel, Shanghao Shi, Ruiquan Huang +2Gradient-Based Attacks Offline Reinforcement Learning
Department of Computer Science, University of Kentucky · Department of Computer Science and Engineering, Washington University in St. Louis
RAIN: Region-Aware Inversion Network for Semantic Watermark Extraction
Sep 14, 2026Zilai LiDiffusion Inversion Watermarking
Independent Researcher
Cascading Gradient Inversion via LT-Code Inspired Peeling in Federated Learning
Sep 10, 2026Saeed Shariati, Mohsen Alambardar MeybodiFederated Learning Gradient-Based Attacks
Department of Applied Mathematics and Computer Science, Faculty of Mathematics and Statistics, University of Isfahan, Isfahan 81746-73441, Iran
Faithful, Sufficient and Understandable: Rethinking Graph Counterfactual Explanations via Discrete Diffusion Inversion
Aug 12, 2026David Bechtoldt, Sidney BenderCounterfactual Explanation Graph Neural Networks
Machine Learning Group, TU Berlin, 10587 Berlin, Germany · BIFOLD–Berlin Institute for the Foundations of Learning and Data, Berlin, Germany
InvFlowFD: Reference-Free and Background-Set-Free Perceptual Music Quality Metric with Flow Matching Inversion
Aug 4, 2026Alon Ziv, Harel Pogoda, Yossi AdiPerceptual Quality Image Quality Assessment
School of Computer Science and Engineering The Hebrew University of Jerusalem, Israel
Dual Inversion for Text-to-Image Diffusion Models: From Both Prompt and Noise Perspectives
Jul 29, 2026Xiaolong Liu, Junjian Li, Yuan Xiao +4Text-To-Image Diffusion Models Text-To-Image
Geely Hangzhou, China · City University of Macau Macau, China
FARI: Robust One-Step Inversion for Watermarking in Diffusion Models
Jul 29, 2026Jindong Yang, Han Fang, Weiming Zhang +2Diffusion Inversion Watermarks
University of Science and Technology of China · 2Anhui Province Key Laboratory of Digital Security
SlerpFlow: Spherical Trajectory Correction for Rectified Flow Inversion
Jul 23, 2026Wenbin Duan, Yan Shu, Zhuoyuan Fu +4Rectified Flow Diffusion Inversion
University of International Relations, Beijing, China · University of Trento, Trento, Italy · Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China.
Feature-Guided Diffusion for Non-Differentiable Inverse Rendering
Jul 19, 2026Andrei-Timotei Ardelean, Michael Fischer, Tim Weyrich +1Diffusion Inversion Inverse Rendering
Friedrich-Alexander-Universität, Erlangen-Nürnberg, Germany · Adobe Research, London, United Kingdom · Charles University, Faculty of Mathematics and Physics, Prague, Czech Republic
Compression Asymmetry and Trajectory Binding in Noise-Anchored Diffusion Inversion
Jul 8, 2026Yongseong Park, Joeun Kim, HoEun Kim +1Diffusion Inversion Inverse Design
Department of EECS DGIST Daegu, Republic of Korea · Department of AI DGIST Daegu, Republic of Korea
Recovering Cloud Microstructures with Cascaded Diffusion Inversion
Jul 6, 2026Hanan Gani, Guy Pulik, Daniel Rosenfeld +2Hyperspectral Image Super-Resolution Generative Super-Resolution
Mohamed Bin Zayed University of Artificial Intelligence · University of California, San Diego · Hebrew University of Jerusalem
When Does High-CFG Diffusion Inversion Fail? A Controlled Study of Prompt--Latent Interactions
Jul 6, 2026Yan Zeng, Yusuke Hosoya, Huyen T. T. Tran +1Diffusion Inversion Diffusion-Based Image Editing
GSIS Tohoku University Sendai, Japan · RIKEN AIP JAPAN
Recovering Input Text from Hidden States: Study of Gradient-Based Inversion of Decoder-Only Language Models
Jul 1, 2026Mikołaj Słowikowski, Maciej Witold MajewskiHidden States Diffusion Inversion
AGH University of Krakow, Faculty of Physics and Applied Computer Science, al. Mickiewicza 30, 30-059 Krakow, Poland
Root-Selecting Fixed-Point Inversion for Rectified Flows via Trajectory Straightness
Jun 16, 2026Semin Kim, Jihwan Yoon, Seunghoon HongRectified Flow Diffusion Inversion
KAIST
Timestep Rescheduling in Diffusion Inversion
Jun 13, 2026Shangquan Sun, Ting Gong, Zhirui Liu +5Diffusion Inversion Timestep
School of CyberScience and Technology, Sun Yat-sen University, Shenzhen Campus, Shenzhen 518107, China · Tsinghua University · CUHK +2
Loss Landscape Diagnosis for Gradient-Based Gray-Scott System Inversion: Disentangling the Roles of PINN Components
Jun 9, 2026Yan YangParametric Physics-Informed Neural Network Diffusion Inversion
Independent Researcher.
Consistent-Inversion: Reverse Consistency Guidance for Structure-Preserving Visual Editing
Jun 5, 2026Xiaocheng Lu, Jingcai Guo, Song GuoDiffusion-Based Image Editing Diffusion Inversion
Hong Kong University of Science and Technology, Hong Kong SAR, China · The Hong Kong Polytechnic University, Hong Kong SAR, China
Training-free image inversion for one-step diffusion models
May 31, 2026Tao Wu, Senmao Li, Yaxing Wang +3Diffusion-Based Image Editing Diffusion Inversion
Computer Vision Center, Universitat Autonoma de Barcelona, Barcelona, 08193, Spain · Mohamed bin Zayed University of Artificial Intelligence (MBZUAI), Abu Dhabi, United Arab Emirates · Jilin University, 130012, Changchun, Jilin, China +3
Stable and Near-Reversible Diffusion ODE Solvers for Image Editing
May 12, 2026Barbora Barancikova, Daniil Shmelev, Cristopher SalviDiffusion-Based Image Editing Diffusion Inversion
Department of Computing, Imperial College London, London, United Kingdom · Department of Mathematics, Imperial College London, London, United Kingdom
PGID: Progressive Guided Inversion and Denoising for Robust Watermark Detection
May 10, 2026Minh Quoc Duong, Chun Tong Lei, Chun Pong LauWatermarks Multi-Bit Watermarking
City University of Hong Kong
A Comparative Study of INDI and NDI with Nonlinear Disturbance Observer for Aerial Robotics
May 7, 2026Benedetta Rota, Mirko Mizzoni, Amr Afifi +2Aerial Robots Diffusion Inversion
Department of Computer, Control and Management Engineering, Sapienza University of Rome, 00185 Rome, Italy · Robotics and Mechatronics group, Faculty of Electrical Engineering, Mathematics, and Computer Science (EEMCS), University of Twente, 7500 AE Enschede, The Netherlands · School of Aerospace, Mechanical, and Mechatronic Engineering (AMME), Faculty of Engineering, University of Sydney, NSW, 2006, Australia
DirectEdit: Step-Level Accurate Inversion for Flow-Based Image Editing
May 4, 2026Desong Yang, Mang YeDiffusion-Based Image Editing Image Editing
National Engineering Research Center for Multimedia Software, School of Computer Science, Wuhan University, Wuhan, China.
Trajectory-Aware Information Matching for Multi-Step Gradient Inversion in Federated Learning
Sep 26, 2025Li Xia, Jing Yu, Zheng Liu +3Federated Learning Diffusion Inversion
Key Laboratory of Ethnic Language Intelligent Analysis and Security Governance of MOE, Minzu University of China, Beijing, 100081, China
Diffusion-aided Task-oriented Semantic Communications with Model Inversion Attack
Jun 24, 2025Xuesong Wang, Mo Li, Xingyan Shi +2Semantic Communications Diffusion Inversion
School of Science and Engineering, The Chinese University of Hong Kong, Shenzhen, Guangdong 518172, China

Sep 24, 2026, cs.LGJ/K move · Enter open · S save

Temporal Gradient Inversion for Private Trajectory Reconstruction in Embodied Reinforcement Learning

Sudip Bhujel, Shanghao Shi, Ruiquan Huang, Ning Zhang+1

Department of Computer Science, University of Kentucky · Department of Computer Science and Engineering, Washington University in St. Louis

Gradient-Based Attacks Offline Reinforcement Learning Diffusion Inversion Gradient

Open paper AskPDF arXiv

Distributed learning in embodied reinforcement-learning agents offers a degree of privacy by retaining raw sensor data on-device and transmitting only policy gradients to the server. Yet temporal structure can amplify this leakage beyond single-frame attacks. We introduce Temporal Reconstruction Attack on Consecutive Encodings (TRACE), an amortized temporal gradient-inversion attack that autoregressively reconstructs the sequence of private observation-action trajectories from per-step policy-learning gradients. The attack exploits two structural signals ignored by prior single-frame methods: (i) cross-time correlation between successive embodied gradients, which we formalize via a conditional mutual-information bound, and (ii) closed-form action recovery from policy-head gradient structure, which we prove exact when standard entropy regularization is sufficiently small. On held-out embodied scenes, TRACE reaches

18.8

dB PSNR with near-perfect action recovery at

3

4.5

ms per reconstructed frame, dominating the learning-based baseline across all reconstruction metrics and exceeding optimization attacks while running orders of magnitude faster. Further evaluation demonstrates TRACE's broader applicability across recurrent, residual, and compact transformer victim architectures, multi-modal inputs, and larger discrete action spaces. Defense experiments suggest that protecting temporal gradient streams may require sequence-aware privacy mechanisms.

Momentum

Related topics

Latest papers 24