Luminous quasars at z  > 6 are key probes of early supermassive black hole (SMBH) growth, massive galaxy evolution, and intergalactic medium properties during cosmic reionization. However, their discovery is very challenging due to their scarcity and overwhelming contamination. Foreground ultracool dwarfs (UCDs) outnumber z  > 6 quasars by 2–4 orders of magnitude. In this work, we leveraged the extensive coverage of DESI Legacy Survey DR10 to conduct a self-supervised search for quasars at z  > 6, directly analyzing multiband optical images and minimizing the biases of the traditional catalog-driven color-color selection criteria. By applying a contrastive learning (CL) method followed by spectral energy distribution (SED) fitting prioritization, we identified 1139 high-priority quasar candidates, for which we expect a competitive 1:1 quasar-to-UCD ratio based on the literature samples. We spectroscopically confirm 16 new quasars at z  = 5.94 − 6.45, achieving a 45% success rate. Remarkably, all 16 objects are relatively bright ( M ₁₄₅₀  < −25.5) quasars, including several with unusual properties such as narrow Ly [[INLINE10]] emission (FWHM ≲ 2600 km s ⁻¹ ), strong Ly [[INLINE13]] +N  V emission with an equivalent width > 100 Å, and a mild observed-frame red near-infrared (NIR) continua ( z  −  J  > 0.4). Notably, three of them would have been missed by traditional color–color selections. These results highlight the power of self-supervised machine learning, combined with SED fitting prioritization, to uncover rare, distant sources beyond the limitations of conventional techniques. Our approach offers a scalable and robust framework for data mining and can be readily extended to forthcoming wide-field surveys such as Rubin/LSST, 4MOST, Euclid , and Roman. These applications will advance the census of high-redshift quasars, potentially extend the redshift frontier, and improve constraints on SMBH formation and evolution in the first billion years of the Universe.