RNAâprotein interactions play essential roles in regulating gene expression. While some RNAâprotein interactions are âspecificâ, that is, the RNAâbinding proteins preferentially bind to particular RNA sequence or structural motifs, others are ânonâRNA specific.â Deciphering the proteinâRNA recognition code is essential for comprehending the functional implications of these interactions and for developing new therapies for many diseases. Because of the high cost of experimental determination of proteinâRNA interfaces, there is a need for computational methods to identify RNAâbinding residues in proteins. While most of the existing computational methods for predicting RNAâbinding residues in RNAâbinding proteins are oblivious to the characteristics of the partner RNA, there is growing interest in methods for partnerâspecific prediction of RNA binding sites in proteins. In this work, we assess the performance of two recently published partnerâspecific proteinâRNA interface prediction tools, PSâPRIP, and PRIdictor, along with our own new tools. Specifically, we introduce a novel metric, RNAâspecificity metric (RSM), for quantifying the RNAâspecificity of the RNA binding residues predicted by such tools. Our results show that the RNAâbinding residues predicted by previously published methods are oblivious to the characteristics of the putative RNA binding partner. Moreover, when evaluated using partnerâagnostic metrics, RNA partnerâspecific methods are outperformed by the stateâofâtheâart partnerâagnostic methods. We conjecture that either (a) the proteinâRNA complexes in PDB are not representative of the proteinâRNA interactions in nature, or (b) the current methods for partnerâspecific prediction of RNAâbinding residues in proteins fail to account for the differences in RNA partnerâspecific versus partnerâagnostic proteinâRNA interactions, or both.