Predicate Quadratic Functional Encryption ( \(\textsf{P}\) - \(\textsf{QFE}\) ) can be seen as an attribute-based extension of Quadratic Functional Encryption ( \(\textsf{AB}\) - \(\textsf{QFE}\) ) that additionally ensures attribute-hiding. In a \(\textsf{P}\) - \(\textsf{QFE}\) scheme, a message pair \((\textbf{x}_1, \textbf{x}_2)\) is encrypted under an attribute vector \(\textbf{a}\) , while the secret keys are generated for predicate vectors \((\textbf{f}, \textbf{p})\) . The decryption outputs the value \((\textbf{x}_1 \otimes \textbf{x}_2)\textbf{f}^\top \) only if the predicate-attribute pair \((\textbf{p}, \textbf{a})\) satisfies a specific linear or quadratic relation. We present the following results: To the best of our knowledge, our constructions advance the state-of-the-art by achieving the first simulation-based security in the AB-QFE setting with the attribute-hiding feature.