265 citations · 1.1k across the 50 of their papers we have counts for
9 papers · 1 filter
Effective Strategies in Zero-Shot Neural Machine Translation
Thanh-Le Ha, Jan Niehues, Alexander Waibel
In this paper, we proposed two strategies which can be applied to a multilingual neural machine translation system in order to better tackle zero-shot scenarios despite not having…
NNLO predictions for dijet production in diffractive DIS
Radek Žlebčík, Daniel Britzger, Jan Niehues +3
Cross sections for inclusive dijet production in diffractive deep-inelastic scattering are calculated for the first time in next-to-next-to-leading order (NNLO) accuracy. These cro…
Transcribing Against Time
Matthias Sperber, Graham Neubig, Jan Niehues +2
We investigate the problem of manually correcting errors from an automatic speech transcript in a cost-sensitive fashion. This is done by specifying a fixed time budget, and then a…
Determination of the strong coupling constant in next-to-next-to-leading order QCD using H1 jet cross section measurements
H1 collaboration, V. Andreev, A. Baghdasaryan +145
The strong coupling constant is determined from inclusive jet and dijet cross sections in neutral-current deep-inelastic scattering (DIS) measured at HERA by the H1…
Comparison of Decoding Strategies for CTC Acoustic Models
Thomas Zenkel, Ramon Sanabria, Florian Metze +4
Connectionist Temporal Classification has recently attracted a lot of interest as it offers an elegant approach to building acoustic models (AMs) for speech recognition. The CTC lo…
Exploiting Linguistic Resources for Neural Machine Translation Using Multi-task Learning
Jan Niehues, Eunah Cho
Linguistic resources such as part-of-speech (POS) tags have been extensively used in statistical machine translation (SMT) frameworks and have yielded better performances. However,…
Analyzing Neural MT Search and Model Performance
Jan Niehues, Eunah Cho, Thanh-Le Ha +1
In this paper, we offer an in-depth analysis about the modeling and search performance. We address the question if a more complex search algorithm is necessary. Furthermore, we inv…
Neural Lattice-to-Sequence Models for Uncertain Inputs
Matthias Sperber, Graham Neubig, Jan Niehues +1
The input to a neural sequence-to-sequence model is often determined by an up-stream system, e.g. a word segmenter, part of speech tagger, or speech recognizer. These up-stream mod…
NNLO QCD corrections to jet production in deep inelastic scattering
James Currie, Thomas Gehrmann, Alexander Huss +1
Hadronic jets in deeply inelastic electron-proton collisions are produced by the scattering of a parton from the proton with the virtual gauge boson mediating the interaction. The…