Rice-Marlin Codes: Tiny and Efficient Variable-to-Fixed Codes
Abstract: Marlin is a Variable-to-Fixed (VF) codec optimized for high decoding speed through the use of small sized dictionaries that fit in the L1 cache of most CPUs. While the size of Marlin dictionaries is adequate for decoding, they are still too large to be encoded fast. We address this problem by proposing two techniques to reduce the alphabet size. The first technique is to encode rare symbols in their own segment, and the second is to combine Marlin dictionaries with Rice encoding, hence our name Rice-Marlin for our new codec. Using those techniques, we are able to reduce the size of Marlin dictionaries by a factor of 16, not only enabling faster encoding speed, but also achieving better compression efficiency.
Paper Prompts
Sign up for free to create and run prompts on this paper using GPT-5.
Top Community Prompts
Collections
Sign up for free to add this paper to one or more collections.