Let p≥5 be a prime and let n be a positive integer. Let Em denote the m-th Euler number, and let Em(x) and Bm(x) denote the Euler and Bernoulli polynomials of degree m, respectively. Z.-W. Sun's conjecture. The following four congruences hold:
n2(n2n)216n(k=0∑np−116k(k2k)2−(p−1)k=0∑n−116k(k2k)2)≡−4p2Ep−3(modp3),
n2(n2n)(n3n)27n(k=0∑np−127k(k2k)(k3k)−(p−3)k=0∑n−127k(k2k)(k3k))≡−23p2Bp−2(31)(modp3),
n2(n2n)(2n4n)64n(k=0∑np−164k(k2k)(2k4k)−(p−2)k=0∑n−164k(k2k)(2k4k))≡−p2Ep−3(41)(modp3),
n2(n3n)(3n6n)432n(k=0∑np−1432k(k3k)(3k6k)−(p−1)k=0∑n−1432k(k3k)(3k6k))≡−20p2Ep−3(modp3).
These conjectures extend known supercongruences for truncated hypergeometric series and connect their p-adic behavior with Euler numbers, Euler polynomials, and Bernoulli polynomials. Their general validity for all primes p≥5 and positive integers n is not established in the supplied source.