Temporal Data Leakage

Follow this topic Ask about this topic

Temporal data leakage studies how information from the future, overlapping records, or later data revisions can distort model training and evaluation. Common safeguards include chronological and patient-level splits, point-in-time feature construction, cutoff-aware retrieval, and audits that distinguish when an event happened from when it became available. Researchers apply these methods across forecasting, clinical prediction, finance, education, and language-model backtesting, using benchmarks and diagnostic tests to measure how leakage changes results. A central tension is preserving enough training data and useful knowledge while enforcing realistic information boundaries; strict evaluation often reduces apparent gains and reveals weak transfer to future periods or new groups.

Latest papers

35 papers6 papers in the last four weeks, up 100% on the four weeks before. 0.0% of all new papers.

The Hidden Look-Ahead: Point-in-Time Leakage in Electricity Market Data
Oct 18, 2026Abdelkrim AlfalahTemporal Data Leakage Electricity Price Forecasting
Goldsmiths, University of London
Hidden Data Challenges in Clinical Machine Learning
Oct 9, 2026Carolin Cissée, Dennis Fast, Raneen Younis +3Training Data Curation Electronic Health Records
Peter L. Reichertz Institute for Medical Informatics of TU Braunschweig and Hannover Medical School, Hannover Medical School, Germany · DATEXIS, Berliner Hochschule für Technik, Germany · Lower Saxony Center for AI and Causal Methods in Medicine (CAIMed), Hannover, Germany +1
Training on the Future: A Delay-Aware Audit of Test-Time Adaptation for Time-Series Forecasting
Oct 8, 2026Mohamed Readh Fentazi, Mazene Ameur, Adlen KsentiniTest-Time Adaptation Time Series Forecasting
Communication Systems Department EURECOM Sophia Antipolis, France
PulseBound: Future-Beat State Forecasting Under an Explicit Information Boundary
Oct 8, 2026Chenyang Xu, Donglin Xie, Xi Xiang +12Predictive Representation Learning Representation Learning
OPPO Health Lab, Guangdong OPPO Mobile Telecommunications Corp., Ltd. · Xidian University · Peking University +3
Repeated-Measure Leakage, Distribution Shift, and Reliability under Partial Observation in Patient World Models
Oct 3, 2026Arjun SubramanianWorld Model Evaluation Data Leakage
Massachusetts Institute of Technology Computer Science and Artificial Intelligence Laboratory (CSAIL)
Frontier Autolab: Organizational Memory, Adversarial Dissent and Temporal Leakage in Multi-Agent LLM Firms Across Fifty Years of Technological Change
Sep 29, 2026Bravish GhoshMulti-Agent LLM Systems LLM Agent Memory
Independent researcher
Time-Series Foundation Models That Understand Data Revisions
Sep 23, 2026Taimoor AhmadTime Series Forecasting Temporal Data Leakage
Superior University Lahore, Pakistan
EHR2Trace: Auditable EHR Data Infrastructure for Patient World Models and Clinical Agents
Sep 18, 2026Xinye Yang, Yuli Wang, Cheng Ting Lin +1Electronic Health Records Data Provenance
Department of Radiology University of Colorado Anschutz School of Medicine Aurora, CO, USA · Department of Radiology Johns Hopkins University School of Medicine Baltimore, MD, USA
The Impossible Trinity of Time-Series Validation: A Conservation Law among Training Sufficiency, Test Coverage, and Temporal Causality
Sep 2, 2026Jiayu LiCross-Validation Temporal Data Leakage
Data Leakage Inflates Generalizability of Power Outage Prediction Models
Aug 25, 2026Yamil Essus, Ranga Raju Vatsavai, Benjamin RachunokOOD Generalization Power Systems
Department of Civil and Mineral Engineering, University of Toronto, Toronto, Canada · Computer Science Department, North Carolina State University, Raleigh, NC · Department of Industrial and Systems Engineering, North Carolina State University, Raleigh, NC +1
KONTOGRAPH: Verified Point-in-Time Feature Consistency and Amortised Explanation for Real-Time Anti-Money Laundering under a 200 ms Decision Budget
Aug 23, 2026Ahmed AbolfadlAnti-Money Laundering Detection Anti-Money Laundering
Faculty of Media Engineering and Technology German University in Cairo Cairo, Egypt
Temporal Leakage in Financial News NLP: A Multi-Architecture Audit with a Regime-Specific M&A Signal
Aug 18, 2026Chenhao Xue, Raslen Guesmi, Siwei Feng +5Temporal Data Leakage Financial Forecasting
Predictive Labs Ltd · University of Oxford · Imperial College London +1
Time-Aware Validation of Machine Learning Fuel Consumption Models: Evidence from 1,Hz Operational Data, CCGS Sir Wilfrid Laurier
Aug 17, 2026Samarasimha Reddy Chittamuru, Ayhan Akinturk, Allison Kennedy +2Maritime Transportation Temporal Data Leakage
National Research Council Canada, Ocean, Coastal and River Engineering, P.O. Box 12093, St. John’s, A1B 3T5, NL, Canada · Department of Computer Science, Memorial University of Newfoundland, 40 Arctic Avenue, St. John’s, A1B 3X5, NL, Canada
Temporal Leakage in LLM Backtesting: Measurement, Validation, and Adjusted Scores
Aug 4, 2026Zeyu Zhang, Bradly C. StadieLLM Evaluation LLM Auditing
Department of Statistics and Data Science Northwestern University
Distilling Temporal Search and Reasoning: Evolving LLMs for Future Prediction via Harness-Assisted Efficient Data Synthesis
Jul 28, 2026Wanxu Cai, Zhengyu Chen, Huaisheng Zhu +3Data Leakage Temporal Reasoning
1The Chinese University of Hong Kong · 2Meituan LongCat Team
HindsightBench: A Black-Box Behavioral Audit Protocol for Parametric Hindsight in Time-Indexed LLM Decision Tasks
Jul 21, 2026Haozhe JiaLLM Auditing Temporal Data Leakage
University College Dublin
Fenced Citation-Context Retrieval for Case Law: Temporal Leakage and Degree Control Across Two Jurisdictions
Jul 19, 2026Yao Liu, Tien-Ping Tan, Zhilan LiuData Leakage Legal IR
The Engineering and Technology College, Chengdu University of Technology, Leshan, China · School of Computer Sciences, Universiti Sains Malaysia, Penang, Malaysia · Department of Art and Design, The Engineering and Technology College, Chengdu University of Technology, Leshan, China
Hindcast: Replaying Prediction Markets to Evaluate LLM Forecasters
Jul 15, 2026Xiao Ye, Jacob Dineen, Evan Zhu +3LLM Evaluation Forecasting Benchmarks
School of Computing and Augmented Intelligence, Arizona State University
Pitfalls of Administrative Censoring in Survival Models with Time-Indexed Inputs
Jul 11, 2026Yanqi Xu, Hui Dai, Carlos Fernandez-Granda +2Data Leakage Survival Analysis
Center for Data Science, New York University · University of Chicagoy · 3NYU Grossman School of Medicine +1
MACROCAST: A Vintage-Consistent Time Series Foundation Model for Real-Time Macroeconomic Forecasting
Jun 27, 2026Andrea Carriero, Davide Pettenuzzo, Shubhranshu ShekharData Leakage Time Series Forecasting
School of Economics and Finance, Queen Mary University of London. · School of Business and Economics, Brandeis University.
Leakage-Aware Benchmarking of LLM Forecasting: Real-Time Nowcasts as the Decision-Time Input for Macro Factor Ranking
Jun 21, 2026Mao Guan, Qian ChenLearning to Rank Look-Ahead Bias in Language Models
Independent Researcher.
WorldReasoner: Evaluating Whether Language Model Agents Forecast Events with Valid Reasoning
Jun 10, 2026Yizhou Chi, Eric Chamoun, Zifeng Ding +1Event Time Prediction LLM Agent Evaluation
Department of Computer Science and Technology University of Cambridge
NumLeak: Public Numeric Benchmarks as Latent Labels in Foundation Models
May 28, 2026Anany KotawalaData Leakage LLM Evaluation
Princeton University, Princeton, NJ, USA.
When Can We Trust Early Warnings? Leakage-Excluded Early Outcome Prediction from LMS Interaction Logs
May 25, 2026Ngoc Luyen Le, Marie-Hélène Abel, Bertrand LaforgeData Leakage AI in Education
Gamaizer, 93340 Le Raincy, France · Université de technologie de Compiègne, CNRS, Heudiasyc (Heuristics and Diagnosis of Complex Systems), CS 60319 - 60203 Compiègne Cedex, France · Sorbonne Université, CNRS UMR 7585, LPMHE (Laboratoire de Physique Nucléaire et des Hautes Énergies), 75252 Paris cedex 05, France
Teaching Large Language Models When Not to Know: Learning Temporal Critique for Ex-Ante Reasoning
May 14, 2026Chenlu Ding, Jiancan Wu, Yanchen Luo +3LLM Prompting LLM Fine-Tuning
University of Science and Technology of China · The Hong Kong Polytechnic University · University of Notre Dame
TEMPO: Temporal Enforcement via Mode-Separated Policy Optimization for Trustworthy LLM Backtesting
May 13, 2026Zeyu Zhang, Bradly C. StadieLLM Evaluation RL for Language Model Reasoning
Department of Statistic and Data Science Northwestern University
Ensuring Reliability in Programming Knowledge Tracing: A Re-evaluation of Attention-augmented Models and Experimental Protocols
May 6, 2026Jaewook Kim, Hyeoncheol KimProgramming Education Temporal Data Leakage
Korea University, Seoul, Republic of Korea
OracleProto: A Reproducible Framework for Benchmarking LLM Forecasting via Knowledge Cutoffs and Temporal Masking
May 5, 2026Yiding Ma, Chengyun Ruan, Kaibo Huang +3LLM Evaluation Forecasting Benchmarks
Beijing University of Posts and Telecommunications
Scaling Point-in-Time Language Models: Enabling Economic Evaluation of Embeddings
Apr 24, 2026Bryan Kelly, Semyon Malamud, Johannes Schwab +1Decoder-Only Language Models Temporal Data Leakage
Yale School of Management · AQR Capital Management · NBER +3
bioLeak: Leakage-Aware Modeling and Diagnostics for Machine Learning in R
Apr 13, 2026Selçuk KorkmazData Leakage ML Reproducibility
Department of Biostatistics and Medical Informatics, Faculty of Medicine, Trakya University, 22030 Edirne, Türkiye

Oct 18, 2026J/K move · Enter open · S save

The Hidden Look-Ahead: Point-in-Time Leakage in Electricity Market Data

Abdelkrim Alfalah

Goldsmiths, University of London

Temporal Data Leakage Electricity Price Forecasting

Open paper AskPDF arXiv

Empirical studies of electricity prices routinely state that their models use only information available at the time of the trading decision. For weather-driven forecasts drawn from commercial archives, this condition is harder to meet than it appears, and whether it holds can change over time. We identify three timing pitfalls. A forecast run is labelled by the issue time of its weather model, but becomes available hours later and keeps changing after publication. Auction gates follow local clock time, while archives use UTC. Outturn series are revised after publication without earlier values being kept. Using the timestamps of 128,656 forecast runs from ten weather-model configurations, matched to the day-ahead and first intraday auction gates in Great Britain over 2022-2026, we find that selecting "the latest run issued before the gate" picks a run not yet published at the gate on more than 99% of delivery days for six of the ten configurations at one of the two gates. For GFS at the day-ahead gate, the share is 0.6% in winter and 99.8% in summer. The exposure also shifts as publication delays drift, and an action margin of a few minutes can cost a full six-hour forecast cycle. The wrongly selected inputs differ materially from those available at the gate, but their apparent accuracy gain ranges from -9% to +24% across variables, models, gates and seasons, so its effect cannot be assumed and must be measured. We argue that point-in-time correctness is a dynamic property of the data pipeline, and propose a protocol that audits it from timestamps before any model is fitted.