We propose a practical model for named entity recognition (NER) that combines word and character-level information with a specific learned representation of the prefixes and suffixes of the word. We apply this approach to multilingual and multi-domain NER and show that it achieves state of the art results on the CoNLL 2002 Spanish and Dutch and CoNLL 2003 German NER datasets, consistently achieving 1.5-2.3 percent over the state of the art without relying on any dictionary features. Additionally, we show improvement on SemEval 2013 task 9.1 DrugNER, achieving state of the art results on the MedLine dataset and the second best results overall (-1.3% from state of the art). We also establish a new benchmark on the I2B2 2010 Clinical NER dataset with 84.70 F-score.

Meson spectroscopy at finite gauge couplingwhereat any perturbative QCD computation would break down -and finite number of colors, from a top-down holographic string model, has thus far been entirely missing in the literature. This paper fills this gap. Using the delocalized type IIA SYZ mirror (with SU ( (Chin Phys C 38:090001, 2014). Through explicit computations, we verify that the vector and scalar meson spectra obtained by the gravity dual with a black hole for all temperatures (small and large) are nearly isospectral with the spectra obtained by a thermal gravity dual valid for only low temperatures; the isospectrality is much closer for vector mesons than scalar mesons. The black-hole gravity dual (with a horizon a e-mail: viitr.dph2015@iitr.ac.in b e-mail: aalokfph@iitr.ac.in c e-mail: krusldph@iitr.ac.in radius smaller than the deconfinement scale) also provides the expected large-N suppressed decrease in vector meson mass with increase of temperature.

We propose an unsupervised strategy for the selection of justification sentences for multihop question answering (QA) that (a) maximizes the relevance of the selected sentences, (b) minimizes the overlap between the selected facts, and (c) maximizes the coverage of both question and answer. This unsupervised sentence selection method can be coupled with any supervised QA approach. We show that the sentences selected by our method improve the performance of a state-of-the-art supervised QA model on two multi-hop QA datasets: AI2's Reasoning Challenge (ARC) and Multi-Sentence Reading Comprehension (MultiRC). We obtain new state-of-the-art performance on both datasets among approaches that do not use external resources for training the QA system: 56.82% F1 on ARC (41.24% on Challenge and 64.49% on Easy) and 26.1% EM0 on MultiRC. Our justification sentences have higher quality than the justifications selected by a strong information retrieval baseline, e.g., by 5.4% F1 in MultiRC. We also show that our unsupervised selection of justification sentences is more stable across domains than a state-of-the-art supervised sentence selection method.

Obtaining the values of the coupling constants of the low energy effective theory corresponding to QCD, compatible with experimental data, even in the (vector) mesonic sector from (the $$ \mathcal{M} $$ M -theory uplift of) a UV-complete string theory dual, has thus far been missing in the literature. We take the first step in this direction by obtaining the values of the coupling constants of the $$ \mathcal{O} $$ O (p4) χPT Lagrangian in the chiral limit involving the NGBs and ρ meson (and its flavor partners) from the $$ \mathcal{M} $$ M -theory/type IIA dual of large-N thermal QCD, inclusive of the $$ \mathcal{O} $$ O (R4) corrections. We observe that ensuring compatibility with phenomenological/lattice results (the values) as given in [1], requires a relationship relating the $$ \mathcal{O} $$ O (R4) corrections and large-N suppression. In other words, QCD demands that the higher derivative corrections and the large-N suppressed corrections in its M/string theory dual, are related. As a bonus, we explicitly show that the $$ \mathcal{O} $$ O (R4) corrections in the UV to the $$ \mathcal{M} $$ M -theory uplift of the type IIB dual of large-N thermal QCD at low temperatures, can be consistently set to be vanishingly small.

Evidence retrieval is a critical stage of question answering (QA), necessary not only to improve performance, but also to explain the decisions of the corresponding QA method. We introduce a simple, fast, and unsupervised iterative evidence retrieval method, which relies on three ideas: (a) an unsupervised alignment approach to soft-align questions and answers with justification sentences using only GloVe embeddings, (b) an iterative process that reformulates queries focusing on terms that are not covered by existing justifications, which (c) a stopping criterion that terminates retrieval when the terms in the given question and candidate answers are covered by the retrieved justifications. Despite its simplicity, our approach outperforms all the previous methods (including supervised methods) on the evidence selection task on two datasets: MultiRC and QASC. When these evidence sentences are fed into a RoBERTa answer classification component, we achieve state-of-the-art QA performance on these two datasets.