A neural compressor turns sentence embeddings into binary codes that retain semantic similarity performance on STS benchmarks while cutting memory by up to 256:1.
Word2Bits - Quantized Word Vectors
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
abstract
Word vectors require significant amounts of memory and storage, posing issues to resource limited devices like mobile phones and GPUs. We show that high quality quantized word vectors using 1-2 bits per parameter can be learned by introducing a quantization function into Word2Vec. We furthermore show that training with the quantization function acts as a regularizer. We train word vectors on English Wikipedia (2017) and evaluate them on standard word similarity and analogy tasks and on question answering (SQuAD). Our quantized word vectors not only take 8-16x less space than full precision (32 bit) word vectors but also outperform them on word similarity tasks and question answering.
fields
cs.IR 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Hamming Sentence Embeddings for Information Retrieval
A neural compressor turns sentence embeddings into binary codes that retain semantic similarity performance on STS benchmarks while cutting memory by up to 256:1.