We present the B-spline Encoded Action Sequence Tokenizer (BEAST), a novel action tokenizer that encodes action sequences into compact discrete or continuous tokens using B-splines. In contrast to ...

Abstract: In this paper, we introduce an Optimized Byte Pair Encoding (OBPE) tokenizer where the algorithm is optimized for the South African languages, including Sesotho, Setswana, Xhosa, Xitsonga, ...

Imagine if you could hide a secret message within a photo, and no one could tell by just looking at it. This is the magic of steganography—a powerful technique that allows us to embed secret ...

End-to-end (E2E) neural networks have emerged as flexible and accurate models for multilingual automatic speech recognition (ASR). However, as the number of supported languages increases, particularly ...

Fast BPE algorithm to generate byte pair encodings from text corpus, it's written in rust and approximately 20x faster than it's python implementation ...