Sparsemap: Differentiable sparse structured inference

V Niculae, A Martins, M Blondel… - … on Machine Learning, 2018 - proceedings.mlr.press
Structured prediction requires searching over a combinatorial number of structures. To
tackle it, we introduce SparseMAP, a new method for sparse structured inference, together …

Discrete latent structure in neural networks

V Niculae, CF Corro, N Nangia, T Mihaylova… - arxiv preprint arxiv …, 2023 - arxiv.org
Many types of data from fields including natural language processing, computer vision, and
bioinformatics, are well represented by discrete, compositional structures such as trees …

Depth-bounding is effective: Improvements and evaluation of unsupervised PCFG induction

L **, F Doshi-Velez, T Miller, W Schuler… - arxiv preprint arxiv …, 2018 - arxiv.org
There have been several recent attempts to improve the accuracy of grammar induction
systems by bounding the recursive complexity of the induction model (Ponvert et al., 2011; …

Bayesian learning for neural dependency parsing

E Shareghi, Y Li, Y Zhu, R Reichart, A Korhonen - 2019 - repository.cam.ac.uk
While neural dependency parsers provide state-of-the-art accuracy for several languages,
they still rely on large amounts of costly labeled training data. We demonstrate that in the …

Training for Gibbs sampling on conditional random fields with neural scoring factors

S Gao, MR Gormley - Proceedings of the 2020 Conference on …, 2020 - aclanthology.org
Most recent improvements in NLP come from changes to the neural network architectures
modeling the text input. Yet, state-of-the-art models often rely on simple approaches to …

Efficient Sampling of Dependency Structures

R Zmigrod, T Vieira, R Cotterell - arxiv preprint arxiv:2109.06521, 2021 - arxiv.org
Probabilistic distributions over spanning trees in directed graphs are a fundamental model of
dependency structure in natural language processing, syntactic dependency trees. In NLP …

Query-focused sentence compression in linear time

A Handler, B O'Connor - arxiv preprint arxiv:1904.09051, 2019 - arxiv.org
Search applications often display shortened sentences which must contain certain query
terms and must fit within the space constraints of a user interface. This work introduces a …

A hitchhiker's guide to efficient non-projective dependency parsing

R Zmigrod - 2022 - repository.cam.ac.uk
Dependency parsing has played a pivotal role in NLP over the last few decades, and non-
projective, graph-based dependency parsers have become a dominant approach for this …

Social Measurement and Causal Inference with Text

KA Keith - 2021 - scholarworks.umass.edu
The digital age has dramatically increased access to large-scale collections of digitized text
documents. These corpora include, for example, digital traces from social media, decades of …

Natural Language Processing for Lexical Corpus Analysis

AK Handler - 2021 - scholarworks.umass.edu
People have been analyzing documents by reading keywords in context for centuries.
Traditional approaches like paper concordances or digital keyword-in-context viewers …