English

Avey-B

Computation and Language 2026-02-18 v1 Artificial Intelligence

Abstract

Compact pretrained bidirectional encoders remain the backbone of industrial NLP under tight compute and memory budgets. Their effectiveness stems from self-attention's ability to deliver high-quality bidirectional contextualization with sequence-level parallelism, as popularized by BERT-style architectures. Recently, Avey was introduced as an autoregressive, attention-free alternative that naturally admits an encoder-only adaptation. In this paper, we reformulate Avey for the encoder-only paradigm and propose several innovations to its architecture, including decoupled static and dynamic parameterizations, stability-oriented normalization, and neural compression. Results show that this reformulated architecture compares favorably to four widely used Transformer-based encoders, consistently outperforming them on standard token-classification and information-retrieval benchmarks while scaling more efficiently to long contexts.

Cite

@article{arxiv.2602.15814,
  title  = {Avey-B},
  author = {Devang Acharya and Mohammad Hammoud},
  journal= {arXiv preprint arXiv:2602.15814},
  year   = {2026}
}
R2 v1 2026-07-01T10:40:18.729Z