Investigating the past occurrences of Yellow River floods provides essential insights into the river's natural variability and recurrent patterns over time. This historical context is indispensable for predicting and mitigating future flood events. However, comprehending the long-term variability of these extreme flood events faces challenges from the limited duration and sparse geographical distribution of gauging station records. Fortunately, flood slackwater deposits (SWDs) within the fluvial stratigraphy provide excellent records for reconstructing extreme floods beyond historical documents and modern observations. Here we scrutinize and synthesize the reported SWD records from the upper and middle Yellow River, and conduct a meta-analysis of these floods, in order to reconstruct the distribution patterns of extreme floods throughout the Holocene. 30 SWD sequences from 57 sites passed our rigorous data quality scrutiny, and subsequently total of 72 flood units (with different age-quality levels) were utilized to reconstruct the spatiotemporal distribution of Holocene extreme floods. Our results identified five extreme flood-rich periods, centered at 8500 yr BP, 6300-6100 yr BP, 4300-4000 yr BP, 3400-3000 yr BP, and 1800-1600 yr BP. The floods in these periods exhibit a significant increase in frequency and a shift in their spatial scale after similar to 5000 yr BP, which are probably modulated by millennial-scale summer insolation, resulting in increased El Nino-Southern Oscillation activity and intensified latitudinal temperature gradient. Our comparison analysis between flood-rich periods and various climatic proxies suggests a centennial-scale 'dry-cold' climatic configuration during these periods. An anomalous low- latitude western North Pacific anticyclone, coupled with an anomalous mid-latitude cyclone-anticyclone pair, contribute to the convergence of water vapor from the western Pacific and Arctic region into northern China. This convergence led to extraordinary rainstorms and extreme floods in the upper and middle Yellow River, which significantly impact the human activity in the lower Yellow River. In addition, identifying Holocene extreme flood-rich periods and their climatic configurations offers new insights for predicting long-term extreme hydrological events in the region. Nonetheless, the uncertainty of our synthesis results owing to the limitation of currently available data should be considered and warrants verification in future studies.