A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
Proposes MorphMine for unsupervised morpheme segmentation to improve word embeddings.
We propose several ways of reusing subword embeddings and other weights in subword-aware neural language models. The proposed techniques do not benefit a competitive character-aware model, but some of them improve the performance of syllable- and morpheme-aware models while showing significant reductions in model sizes…
Deep networks respond to specific linguistic units, not arbitrary patterns.
New models improve morpheme segmentation in low-resource languages.
We describe a simple neural language model that relies only on character-level inputs. Predictions are still made at the word-level. Our model employs a convolutional neural network (CNN) and a highway network over characters, whose output is given to a long short-term memory (LSTM) recurrent neural network language mo…
Proposes RDASS for better Korean text summarization evaluation.