The Flow has not summarised this story yet — read it at arXiv Computation and Language.
arXiv:2609.09691v1 Announce Type: new Abstract: When training data are limited, increasing parameter count is not the only way to improve language-model performance. A small parameter set, when repea...
arXiv:2606. 18811v1 Announce Type: cross Abstract: Learned sparse retrieval (LSR) models such as SPLADE have traditionally used BERT-style masked language models as backbone encoders.
arXiv:2012. 02110v2 Announce Type: replace-cross Abstract: Pre-trained language models have significantly advanced natural language processing (NLP), especially with the introduction of BERT and its optimized version, RoBERTa.
arXiv:2606. 25331v1 Announce Type: cross Abstract: Modern large language models are predominantly trained with autoregressive factorization and causal attention.
arXiv:2609.13154v1 Announce Type: new Abstract: Recent advances in large language models (LLMs) have made prompts increasingly large and complex. Techniques such as chain-of-thought reasoning (Wei et...
arXiv:2609.18156v1 Announce Type: new Abstract: Teochew has a substantial speaker community and exhibits distinctive lexical, syntactic, and pragmatic features, yet textual resources for evaluating l...