Syntax inference (inferring protocol syntaxes) is the process in which protocol field boundaries, offset locations, and endianness are inferred.
Yurichev  provides basic and important fundamentals such as integral datatypes, signed number representations, AND, AND, and OR as subtraction and addition, XOR (exclusive OR), Population Count, Endianness, Memory, CPU, and Hash functions towards reverse engineering.
The extension on BIL mainly involves two aspects: 1) the arguments indicating the endianness
in store and load instructions are omitted, since currently we focus on x86 platform; 2) call instruction is introduced to provide the knowledge of the semantics of platform specific functions.
However, the population was split quite evenly between L and B, with a small minority of Bits who still remembered the single-tape Turing machine and preferred unary encoding (U), without any of this endianness
IRI's ability to maintain and change the endianness of data means that files can be correctly processed on big, little, or bi-endian hardware.
IRI will first deliver endianness support in three of its current tools:
CoSort: Inputs, processes, and outputs data as big or little endian regardless of the endianness of the execution platform.
Further performance gains could be achieved if the code were to manage details such as endianness