About this paper
A market-conditioned joint-embedding predictive architecture improves financial time-series prediction by conditioning on a leave-one-out market factor, enhancing market exposure, risk, liquidity, and return forecasts.
Much of the variation in financial time series is driven by market-wide shocks that cannot be predicted from an asset's own history. Per-asset self-supervised learning treats this variation as irreducible noise. Under a latent factor model, we show that conditioning prediction on the realized market factor makes the asset's market exposure necessary for optimal prediction, while reducing the noise against which the asset-specific component is learned. Building on this insight, we introduce Market-Conditioned JEPA (MC-JEPA), a joint-embedding predictive architecture for financial time series. During training, we construct a leave-one-out market condition for each asset from the future…
Market-Conditioned JEPAMC-JEPAjoint-embedding predictive architectureself-supervised learningper-asset JEPAlatent factor modelmarket factorleave-one-out market conditionfuture target embeddingsfactor loadings
- Upvotes
- 4
- Authors
- 6
- Published
- Nov 10, 2026
Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.
Trending history
Market-Conditioned JEPA: Learning from Unpredictable Market Shocks was first seen on Hugging Face · Daily papers on Nov 11, 2026. Since then it has appeared in 19 checks on 3 different days. Its best position was #21 of 30, reached on Nov 11, 13:28 UTC. It dropped off the list after Nov 13, 03:28 UTC.