He and colleagues: momentum contrast for visual representation learning

What changed

Used a momentum-updated encoder and queue of image representations to train contrastive visual features independently of very large batches.

Scientific scope

Used a momentum-updated encoder and queue of image representations to train contrastive visual features independently of very large batches. Scope limitation: First preprint was 2019; year follows CVPR 2020. MoCo builds on prior contrastive learning and is distinct from later MoCo versions.

Evidence

The retained primary-source abstract, bibliographic record and stated review scope support this provisional assessment. No independent experimental replication by WRS is claimed. First preprint was 2019; year follows CVPR 2020. MoCo builds on prior contrastive learning and is distinct from later MoCo versions.

Contributors

• The named authors listed in the linked research record · SCIENCE: Complete source byline retained in the research record; human research direction, method development and evaluation. Individual labor shares are not measured.

• He and colleagues: momentum contrast for visual representation learning — original paper
• He and colleagues: momentum contrast for visual representation learning — official company source

Provisional WRS assessment, edition 2026-10-16.1. Independent expert calibration is pending. This grade does not certify correctness or assign ownership.

Full scientific record · Browse contribution grades · Request a correction