Data Leakage

Follow this field Ask about this field

Recent momentum

emerging

0 papers in the last 28 days · 0.0% of indexed attention

Twelve weeks of publication activity for this topic as it is defined today.

Weekly history

Recent digests

What was published in this field, kept on the site without email delivery.

Period ending 2026-09-21

6 new papers

A weekly snapshot of new work published in Data Leakage.

Inside this field

Focused directions

Focused topic

Data Leakage

Within Data Leakage

85 papersExplore

Focused topic

Membership Inference Attacks

Within Data Leakage

59 papersExplore

Focused topic

Leakage

Within Data Leakage

49 papersExplore

Focused topic

Membership Inference

Within Data Leakage

27 papersExplore

198 papers

Latest in Data Leakage

Browse all topics →

idSCD: Identifying Training Datasets through Semantic Correlation Descriptors
May 28, 2026Andrada Gobeaja, Ionut Hodoroaga, Elena Burceanu +1Membership Inference Cross-Dataset Benchmark
NumLeak: Public Numeric Benchmarks as Latent Labels in Foundation Models
May 28, 2026Anany KotawalaLeakage Foundation Model
Entity-Collision: A Stratified Protocol for Attributing Retrieval Lift in Agent Memory
May 28, 2026Youwang DengAgent Memory Factual Recall
FPLIER: Federated Pathway-Level Information Extractor
May 28, 2026Daniele Malpetti, Christian Berchtold, Francesco Gualdi +3Privacy-Preserving Framework Spatial Transcriptomics
A Full-Pipeline Framework for Evaluating Membership Inference Attacks in Machine Learning
May 28, 2026Ding Chen, Xinwen Cheng, Xuyang Zhong +3Membership Inference Attacks Machine Learning
Auditing Training Data in Generative Music Models via Black-Box Membership Inference
May 28, 2026Yi Chen Liu, Jiawei Yu, Kexin Cao +3Music Generation Membership Inference
Reasoning that Travels: Dissecting How Chain-of-Thought Transfers Across Models
May 27, 2026Xinyuan Cheng, Beiduo Chen, Philipp Mondorf +1Reasoning Traces Efficient Chain-Of-Thought Reasoning
When Helpful Context Leaks: Privacy Risks in Domain-Adapted ASR
May 27, 2026Maike Züfle, Jan NiehuesData Leakage AI Privacy Risks and Protection
MRMMIA: Membership Inference Attacks on Memory in Chat Agents
May 27, 2026Kai Chen, Yan Pang, Tianhao WangMembership Inference Attacks Agent Memory
Detectability in Diversity: Improved Canary Crafting for Privacy Auditing in One Run
May 26, 2026Mathieu Dagréou, Aurélien BelletEfficient Privacy Loss Membership Inference Attacks
Black-box Membership Inference Attacks on the Pre-training Data of Image-generation Models
May 26, 2026Tao Qi, Huili Wang, Yuanhong Huang +6Membership Inference Attacks Model Pretraining
MinhwaNet: Faithful but Insufficient Object Grounding in Korean Folk Painting
May 26, 2026Joonhyung BaeKorean Representation Quality
AnchorDiff: Training-Free Concept Grounding for MM-DiTs via Anchor-Based Graph Propagation
May 26, 2026Jian Zhang, Zhijun ZhangMultimodal Diffusion Transformers Vision-Language Model Grounding
Vectors Are Not Neutral: Sensitive-Information Inference from Exported LLM Representations in Summarization
May 26, 2026Weixin Liu, Bowen Qu, Juming Xiong +3Summarization Data Leakage
On Reliability of Efficient Membership Inference Vulnerability Evaluation
May 25, 2026Joonas Jälkö, Gauri Pradhan, Ossi Räisä +1Membership Inference Attacks False Positive Rate
When Can We Trust Early Warnings? Leakage-Excluded Early Outcome Prediction from LMS Interaction Logs
May 25, 2026Ngoc Luyen Le, Marie-Hélène Abel, Bertrand LaforgeEarly Warning Leakage
Testing the Test: Score-Direction Instability in Class-Split Anomaly Detection
May 23, 2026Alejandro Ascarate, Leo Lebrat, Rodrigo Santa Cruz +2Anomaly Score Anomaly Detection
Boundary-targeted Membership Inference Attacks on Safety Classifiers
May 21, 2026Anthony Hughes, Alexander Goldberg, Prince Jha +3Membership Inference Attacks Safety
Pretraining Data Exposure in Large Language Models: A Survey of Membership Inference, Data Contamination, and Security Implications
May 21, 2026Ziyi Tong, Feifei Sun, Le Minh NguyenMembership Inference Large Language Model Pretraining
DISC: Decoupling Instruction from State-Conditioned Control via Policy Generation
May 20, 2026Hanxiang Ren, Pei Zhou, Xunzhe Zhou +1Language-Conditioning Space Policy Visuomotor Policy
An Application-Layer Multi-Modal Covert-Channel Reference Monitor for LLM Agent Egress
May 20, 2026Alfredo MetereLanguage Model Evasion Attacks Leakage
Auditing Privacy in Multi-Tenant RAG under Account Collusion
May 19, 2026Florian A. D. BurnatEfficient Privacy Loss Algorithm Auditing
Shaping the Prior: How Synthetic Task Distributions Determine Tabular Foundation Model Quality
May 18, 2026Mohamed Bouadi, Nassim Bouarour, Varun Kulkarni +3Tabular Foundation Models Prior Knowledge
AMR-SD: Asymmetric Meta-Reflective Self-Distillation for Token-Level Credit Assignment
May 18, 2026Zhenlin Wei, Pu Jian, Yingzhuo Deng +6Unsupervised On-Policy Self-Distillation Token-Level Supervision
Single-Sample Black-Box Membership Inference Attack against Vision-Language Models via Cross-modal Semantic Alignment
May 17, 2026Jiaqing Li, Yajuan Lu, Xiaochuan Shi +3Membership Inference Attacks Cross-Modal Alignment
Privacy Policy Enforcement Guardrails for Data-Sensitive Retrieval-Augmented Generation
May 16, 2026Osama Zafar, Alexander Nemecek, Yiqian Zhang +5AI Privacy Risks and Protection Data Leakage
PrivScope: Task-scoped Disclosure Control for Hybrid Agentic Systems
May 15, 2026Shafizur Rahman Seeam, Zhengxiong Li, Zhiyuan Yu +3Data Leakage Privilege Illusion
Membership Inference Attacks on Discrete Diffusion Language Models
May 15, 2026Shailesh KasivelrajanMasked Diffusion Language Models Membership Inference Attacks
ReMIA: a Powerful and Efficient Alternative to Membership Inference Attacks against Synthetic Data Generators
May 14, 2026Davide Scassola, Andrea Coser, Sebastiano SaccaniMembership Inference Attacks Synthetic Data
Privacy Evaluation of Generative Models for Trajectory Generation
May 14, 2026Stavros Bouras, Ioannis Kontopoulos, Chiara Pugliese +5Generative Models Latent Trajectories
Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning
May 14, 2026Chenlu Ding, Jiancan Wu, Yanchen Luo +3Leakage
LoREnc: Low-Rank Encryption for Securing Foundation Models and LoRA Adapters
May 13, 2026Beomjin Ahn, Jungmin Kwon, Chanyong Jung +1Low-Rank Adaptation Adapters Wireless Foundation Models
DistractMIA: Black-Box Membership Inference on Vision-Language Models via Semantic Distraction
May 12, 2026Hongyi Tang, Zhihao Zhu, Yi YangMembership Inference Membership Inference Attacks
FERMI: Exploiting Relations for Membership Inference Against Tabular Diffusion Models
May 12, 2026Abtin Mahyar, Masoumeh Shafieinejad, Yuhan Liu +1Membership Inference Attacks Membership Inference
Unlearning with Asymmetric Sources: Improved Unlearning-Utility Trade-off with Public Data
May 11, 2026Ahmed Mehdi Inane, Vincent Quirion, Gintare Karolina Dziugaite +1Unlearning Aims Differential Privacy
Can You Keep a Secret? Involuntary Information Leakage in Language Model Writing
May 11, 2026Ari Holtzman, Peter WestData Leakage Leakage
PRISM: Generation-Time Detection and Mitigation of Secret Leakage in Multi-Agent LLM Pipelines
May 11, 2026Riya Tapwal, Abhishek Kumar, Carsten MapleData Leakage Language Model Evasion Attacks
Security Risks in Tool-Enabled AI Agents: A Systematic Analysis of Privileged Execution Environments
May 10, 2026Hardik GoelArtificial Intelligence Safety Artificial Intelligence Agents
Neural Information Causality
May 10, 2026Jeongho Bang, Marcin PawłowskiInformation-Theoretic Limits Causal Representation Learning
BetaEdit: Null-Space Constrained Sequential Model Editing
May 10, 2026Bingqing Liu, Wei Liu, Yuhua LiMultimodal Knowledge Editing Sequence Modeling
Generating Leakage-Free Benchmarks for Robust RAG Evaluation
May 9, 2026Jiayi Liu, Jiaxing Zhang, Bowen Jin +1Retrieval-Augmented Generation Systems Reasoning Graphs
AgentCollabBench: Diagnosing When Good Agents Make Bad Collaborators
May 9, 2026Aritra Mazumder, Shubhashis Roy Dipta, Nusrat Jahan Lia +10Multi-Agent Systems Data Leakage
SHRED: Retain-Set-Free Unlearning via Self-Distillation with Logit Demotion
May 8, 2026Zizhao Hu, Ameya Godbole, Johnny Tian-Zheng Wei +3Large Language Model Unlearning Logit Lens
On Privacy Leakage in Tabular Diffusion Models: Influential Factors, Attacker Knowledge, and Metrics
May 7, 2026Masoumeh Shafieinejad, D. B. Emerson, Behnoosh Zamanlooy +5Efficient Privacy Loss Synthetic Tabular Data
PACZero: PAC-Private Fine-Tuning of Language Models via Sign Quantization
May 7, 2026Murat Bilgehan Ertan, Xiaochen Zhu, Phuong Ha Nguyen +2Parameter-Efficient Fine-Tuning Membership Inference Attacks
The frame-level leakage trap: rethinking evaluation protocols for intrinsic image decomposition, with source-separable uncertainty as a case study
May 7, 2026Jihwan WooPixel-Level Prediction Scene Representation
Metonymy in vision models undermines attention-based interpretability
May 7, 2026Ananthu Aniraj, Cassio F. Dantas, Dino Ienco +2Model Interpretability Vision Foundation Models
LeakDojo: Decoding the Leakage Threats of RAG Systems
May 7, 2026Maosen Zhang, Jianshuo Dong, Boting Lu +4Retrieval-Augmented Generation Systems Language Model Evasion Attacks
From Beats to Breaches:How Offensive AI Infers Sensitive User Information from Playlists
May 6, 2026Stefano Cecconello, Mauro Conti, Luca Pajola +2Membership Inference Attacks Artificial Intelligence Systems
Membership Inference Attacks for Retrieval Based In-Context Learning for Document Question Answering
May 5, 2026Tejas Kulkarni, Antti Koskela, Laith ZumotMembership Inference Attacks In-Context Learning
PIIGuard: Mitigating PII Harvesting under Adversarial Sanitization
May 4, 2026Mingshuo Liu, Yiwei Zha, Min ChenPrompt-Level Mitigations Language Model Evasion Attacks
On the Privacy of LLMs: An Ablation Study
May 4, 2026Karima Makhlouf, Lamiaa Basyoni, Syed Khaderi +4Membership Inference Attacks Language Model Evasion Attacks
Stochastic Modeling of Human-Machine Authentication Channels under Partial Information Leakage
May 3, 2026Nilesh Chakraborty, Mohammad Zulkernine, Burak KantarciAuthentication Data Leakage
Metric-Normalized Posterior Leakage (mPL): Attacker-Aligned Privacy for Joint Consumption
May 1, 2026Gaoyi Chen, Minghao Li, Weishi Shi +4Differential Privacy AI Privacy Risks and Protection
Most Current Model Organisms Are Leaky: Perplexity Differencing Often Reveals Finetuning Objectives
May 1, 2026Mohammed Abu Baker, Luca Baroni, Dan WilhelmHarmful Fine-Tuning Large Language Model Safety
E-MIA: Exam-Style Black-Box Membership Inference Attacks against RAG Systems
May 1, 2026Zelin Guan, Shengda Zhuo, Zeyan Li +4Membership Inference Membership Inference Attacks
Revealing the Impact of Visual Text Style on Attribute-based Descriptions Produced by Large Visual Language Models
Apr 30, 2026Xiaomeng Wang, Martha Larson, Zhengyu ZhaoLarge Vision Language Models Natural Language Descriptions
Secret Stealing Attacks on Local LLM Fine-Tuning through Supply-Chain Model Code Backdoors
Apr 30, 2026Zi Li, Tian Zhou, Wenze Li +3Language Model Evasion Attacks Large Language Model Fine-Tuning
Path-Lock Expert: Separating Reasoning Mode in Hybrid Thinking via Architecture-Level Separation
Apr 29, 2026Shouren Wang, Wang Yang, Chuang Ma +7Mixture-Of-Expert Language Models Hybrid
Quantamination: Dynamic Quantization Leaks Your Data Across the Batch
Apr 29, 2026Hanna Foerster, Ilia Shumailov, Cheng Zhang +3Extreme Low-Bit Quantization Quantizer

May 28, 2026 · cs.LGJ/K move · Enter open · S save

idSCD: Identifying Training Datasets through Semantic Correlation Descriptors

Andrada Gobeaja, Ionut Hodoroaga, Elena Burceanu, Marius Leordeanu

POLITEHNICA University of Bucharest · Bitdefender, Romania · Institute of Mathematics of the Romanian Academy

Can a dataset be recognized from the spurious correlations it induces during training? We argue that datasets leave dataset-specific traces in a model's learned semantic correlation structure: incidental regularities that are predictive within a dataset, but not causal for the underlying task, can be internalized during training. We use this insight to study dataset-level membership inference, moving beyond existing methods that rely on behavioral or distributional evidence such as confidence scores, losses, margins, generated samples, or query responses. We introduce a white-box semantic fingerprinting approach based on semantic correlation descriptors (SCDs), which capture the semantic correlation structure learned by a model and make it comparable across dataset mixtures. In a controlled leave-one-dataset-out diagnostic, SCDs recover dataset-specific changes and perfectly separate matching from non-matching dataset pairs. We then propose a practical SCD-based membership score that tests whether a target dataset is part of a model's training mixture using only the model's SCD and the target dataset's standalone SCD, without requiring leave-one-dataset-out models. Across three diverse experimental settings, with dataset groups for natural language inference, emotion classification, and medical text classification, we test both the advantages and limitations of SCD-based membership inference with different degrees of semantic separation and keyword support between dataset splits. On average, the classifier based on this score achieves the highest performance and the lowest std, outperforming black-box baselines RMIA, Attack-P, and LiRA, as well as the white-box SIF baseline. These results show that dataset membership can be traced through internal semantic correlations, with the largest relative gain exceeding 60% in ROC-AUC when dataset groups expose distinct semantic particularities.

Membership Inference Cross-Dataset Benchmark Training Data

Open paper Ask about it PDF