Data Fusion

Follow this topic Ask about this topic

Latest papers

24 papers3 papers in the last four weeks, against 1 the four weeks before. 0.0% of all new papers.

Benign Overfitting under Heterogeneous Input Fusion
Oct 7, 2026Houzhen Liu, Xiaobo XiaBenign Overfitting Statistical Learning Theory
University of Science and Technology of China
Data Fusion for Errors-in-Variables
Oct 5, 2026Huali Zhao, Molei Liu, Tianying WangData Fusion Parameter Estimation
School of Mathematics and Statistics, Huazhong University of Science and Technology · Department of Biostatistics, Peking University Health Science Center; Beijing International Center for Mathematical Research, Peking University · Department of Statistics, Colorado State University
When Satellite and Street Views Disagree: Cross-View Complementarity in Disaster Damage Assessment
Oct 3, 2026Yifan Yang, Lei ZouDisaster Damage Assessment Building Damage Assessment
Department of Geography, Texas A&M University, College Station, Texas, USA
Cross-Block Conditioning in Deep Boltzmann Machines for Statistical Data Fusion
Sep 14, 2026Junichiro NiimiLearning with Missing Data Conditional Distribution Estimation
Meijo University
When More Foundation Models Means Less: Diagnosing and Addressing Multi-View Fusion Failure
Aug 18, 2026Yibo Liu, Bowen JiangMulti-View Feature Fusion Multi-View Learning
Hainan Campus, Beijing University of Posts and Telecommunications Lingshui, Hainan, China
Fusion Anything: A Generalized Multimodal Foundation Model
Aug 16, 2026Huizi Cui, Zongbo Han, Chenggong Ding +6Unified Multimodal Models Multimodal Pretraining
Tianjin University · Beijing University of Posts and Telecommunications
Finding Change in Satellite Archives from Text: How to Combine Before-and-After Images Efficiently
Jul 30, 2026Simon Roy, Mark Bong, Giovanni BeltrameCross-Modal Retrieval Feature Fusion
Polytechnique Montréal, Montréal, QC, Canada
Extreme Volatility Warning under Label Scarcity via Multi-Source Anomaly Fusion
Jul 26, 2026Jin Qian, Zhangzhi Xiong, Mingrui Li +1Multimodal Anomaly Detection Semi-Supervised Anomaly Detection
ShanghaiTech University, Shanghai, China
Characterisation of Density-based FM generation methods in the context of Information Fusion
Jul 25, 2026Yanhao Huang, Christian WagnerUncertainty Quantification Fuzzy Logic
Lab for Uncertainty in Data and Decision Making (LUCID) School of Computer Science, University of Nottingham Nottingham, United Kingdom
Segregate, Refine, Integrate: Decomposing Multimodal Fusion for Sentiment Analysis
Jul 14, 2026Alexios Filippakopoulos, Elias Kallioras, Nikolaos Xiros +2Cross-Modal Representation Learning Multimodal Large Language Models
National Technical University of Athens, Greece · Athena Research Center, Greece · University of Bern, Switzerland +2
SHAP-Weighted Cross-Modal Expert Fusion for Emotion and Sentiment Recognition: Evidence and Limits
Jul 9, 2026Adis Alihodzic, Selma Skopljakovic HubljarMixture-of-Experts Models Multimodal Emotion Recognition
Department of Mathematical and Computer Sciences, Faculty of Science, University of Sarajevo, Sarajevo, Bosnia and Herzegovina.
MaDI-Bench: An End-to-End Data Integration Benchmark
Jun 29, 2026Aaron Steiner, Ralph Peeters, Christian BizerBenchmark Design Data Integration
Data and Web Science Group University of Mannheim Mannheim, Germany
Single and Multi Truth Data Fusion using Large Language Models
Jun 26, 2026Hira Beril Kucuk, Norman W Paton, Jiaoyan Chen +1LLM Prompting Data Fusion
Department of Computer Science University of Manchester, Oxford Road, Manchester M13 9PL, UK
LCAi: Life Cycle Assessment with big data fusion and retrieval-augmented generation-assisted interpretation
Jun 25, 2026Georgios Tsironis, Juan D. Medrano-Garcia, Gonzalo Guillen-GosalbezRetrieval-Augmented Generation LLM Grounding
Institute for Chemical and Bioengineering, Department of Chemistry and Applied Biosciences, ETH Zürich, Vladimir Prelog Weg 1, Zürich 8093, Switzerland · NCCR Catalysis, Zurich 8093, Switzerland
When Does Quality-Aware Multimodal Fusion Matter? A Leakage-Safe Diagnostic for Decision-Level Dependence
Jun 25, 2026Jaden Moon, Arvind Pillai, Andrew CampbellUncertainty-Aware Multimodal Fusion Multimodal Fusion
Dartmouth College, United States
Compositional Boundaries for Density Fusion
Jun 4, 2026Ratan Bahadur Thapa, Ali Darijani, Jürgen Beyerer +1Data Fusion
Plug-and-Play Logit Fusion for Heterogeneous Pathology Foundation Models
Apr 9, 2026Gexin Huang, Anqi Li, Yusheng Tan +4Computational Pathology Ensemble Learning
University of British Columbia, Vancouver, BC, Canada · Vector Institute, Toronto, ON, Canada · Washington University in St. Louis, St. Louis, MO, USA
Document Classification Pattern Recognition via Information Fusion: A Systematic Review of Multimodal and Multiview Representation Approaches
Apr 7, 2026Marcin Michał MirończukData Fusion Multimodal Fusion
National Information Processing Institute, al. Niepodległości 188b, 00-608, Warsaw, Poland
Calibrated Fusion for Heterogeneous Graph-Vector Retrieval in Multi-Hop QA
Mar 30, 2026Andre BacellarHybrid Retrieval Data Fusion
Rethinking Multimodal Fusion for Time Series: Text Modalities Need Constrained Fusion
Mar 23, 2026Seunghan Lee, Jun Seo, Jaehoon Lee +7Cross-Modal Learning Time Series Forecasting
LG AI Research Seoul, South Korea
Automatic End-to-End Data Integration using Large Language Models
Mar 11, 2026Aaron Steiner, Christian BizerData Integration Data Fusion
Data and Web Science Group, University of Mannheim, Mannheim, Germany
Estimation and Optimization of Ship Fuel Consumption in Maritime: Review, Challenges and Future Directions
Feb 25, 2026Dusica Marijan, Hamza Haruna Mohammed, Bakht ZamanMaritime Transportation Data Fusion
Simula Research Laboratory, Kristian Augusts 23, Oslo, 0164, Norway
Enhancing Diversity and Feasibility: Joint Population Synthesis from Multi-source Data Using Generative Models
Feb 17, 2026Farbod Abbasi, Zachary Patterson, Bilal FarooqSynthetic Population Generation Data Fusion
Concordia University, Montreal, QC, Canada · Concordia Institute for Information Systems Engineering, Concordia University, Montreal, QC, Canada · Toronto Metropolitan University, Toronto, ON, Canada
Closing Gaps in Emissions Monitoring with Climate TRACE
Nov 24, 2025Brittany V. Lancellotti, Jordan M. Malof, Aaron Davitt +32Data Fusion
Nicholas Institute for Energy Environment & Sustainability, Duke University, Durham, NC, USA. · Department of Electrical Engineering and Computer Science, University of Missouri, Columbia, MO, USA. · WattTime, Oakland, CA, USA. +7

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

Benign Overfitting under Heterogeneous Input Fusion

Houzhen Liu, Xiaobo Xia

University of Science and Technology of China

Benign Overfitting Statistical Learning Theory Data Fusion Multimodal Fusion

Open paper AskPDF arXiv

Benign overfitting is extensively studied when learning from a single high-dimensional input, but its behavior under heterogeneous input fusion remains largely unexplored. We study this question for minimum-norm linear interpolation under a heterogeneous Gaussian design, comparing two statistically dependent input blocks with their fusion while holding the underlying population task fixed. For regression, we identify a full-spectrum covariance certificate whose asymptotic status is independent of the cutoff threshold and prove that it is preserved by every positive-semidefinite joint covariance consistent with the two marginals. This protection is sharp, yet it does not extend to all benign regression problems: outside the certified regime, two benign marginals can have a harmful fusion. For one-sparse Gaussian classification, benignity in the regular regime is characterized by the balance between surviving predictive signal and nuisance contamination. Fusion can move these two quantities in opposite directions, and within this model class every marginal-to-joint benign/non-benign pattern is attainable. We further show that the same fused input can have qualitatively different effects on regression and classification. These results establish that benign overfitting under heterogeneous fusion is determined by the joint signal and spectral geometry created by input interaction, rather than by marginal benignity alone.