The IWSLT 2016 Evaluation Campaign

By Mauro Cettolo, Niehues Jan, Stüker Sebastian, Luisa Bentivogli, Roldano Cattoni, Marcello Federico. 2016

98 citations recorded by the source corpus.

Subfields: Natural Language Processing, Evaluation & Benchmarks

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Natural Language Processing

In Evaluation & Benchmarks

Citations in the corpus

The Natural Language Decathlon: Multitask Learning as Question Answering (2018, methodology): “We use the 2016 English to German training data prepared for the International Workshop on Spoken Language Translation (IWSLT) [Cettolo et al., 2016].”
Exploring Versatile Generative Language Model Via Parameter-Efficient Transfer Learning (2020, methodology): “Neural Machine Translation (NMT) We use the spoken German-English translation dataset IWSLT (Cettolo et al., 2016) as our NMT benchmark.”
On Adversarial Examples for Character-Level Neural Machine Translation (2018, methodology): “Using this architecture, the BLEU scores on our datasets are competitive with the submissions to the IWSLT (Mauro et al., 2016).”
Stronger Baselines for Trustable Results in Neural Machine Translation (2017, background): “Neural models have consistently shown top performance in shared evaluation tasks (Bojar et al., 2016; Cettolo et al., 2016) and are becoming the technology of choice for commercial MT service providers (Wu et al.”
Domain Adaptive Inference for Neural Machine Translation (2019, methodology): “We adapt first to the IWSLT 2016 TED task (Cettolo et al., 2016), and then sequentially to the APE 2017 IT task (Turchi et al., 2017).”
BP-Transformer: Modelling Long-Range Context via Binary Partitioning (2019, methodology): “We conduct experiments with TED Talks Chinese-to-English(Zh-En) dataset from IWSLT 2014 and 2015 (Cettolo et al., 2012, 2016), the average document length is 120 (in sentences).”
One Size Does Not Fit All: Comparing NMT Representations of Different Granularities (2019, methodology): “Data and Languages We trained NMT systems for four language pairs: German-English, Czech-English, Russian-English, and English-German, using data made available through two popular machine translation campaigns, namely, WMT (Bojar et al., 2017) and IWSLT (Cettolo et al., 2016).”
Overview of the IWSLT 2017 Evaluation Campaign (2017, methodology): “For each language direction, the output of the selected 9 systems on the HE set was assigned to professional translators to be post-edited (for all the details about data preparation and postediting see [20, 21, 22]).”
On the Linguistic Representational Power of Neural Machine Translation Models (2019, unknown): “2017) and IWSLT (Cettolo et al. 2016).”
Learning Hidden Unit Contribution for Adapting Neural Machine Translation Models (2018, result): “Similar to Freitag and Al-Onaizan (2016), we present results on the IWSLT 2016 2 German to English TED dataset (Cettolo et al., 2016), consisting of transcribed and translated TED talks.”
FINDINGS OF THE IWSLT 2020 EVALUATION CAMPAIGN (2020, unknown): “…conference (Akiba et al., 2004; Eck and Hori, 2005; Paul, 2006; Fordyce, 2007; Paul, 2008, 2009; Paul et al., 2010; Federico et al., 2011, 2012; Cettolo et al., 2013, 2014, 2015, 2016, 2017; Niehues et al., 2018, 2019) for the study, development and evaluation of spoken language translation…”
Monotonic Multihead Attention (2019, background): “IWSLT15 English-Vietnamese TED talks from IWSLT 2015 Evaluation Campaign (Cettolo et al., 2016).”
Findings of the 2017 DiscoMT Shared Task on Cross-lingual Pronoun Prediction (2017, methodology): “…task on cross-lingual pro-noun prediction, the TED training and development sets come from either the MT tasks of the IWSLT evaluation campaigns (Cettolo et al., 2016) or from past editions of the task (Hardmeier et al., 2015; Guillou et al., 2016); the test sets are built from 16 TED talks…”
A Tree-based Decoder for Neural Machine Translation (2018, methodology): “…1) the KFTT (ja-en) dataset (Neubig, 2011), which consists of Japanese-English Wikipedia articles; 2) the IWSLT2016 German-English (de-en) dataset (Cettolo et al., 2016), which consists of TED Talks transcriptions; and 3) the LORELEI Oromo-English (or-en) dataset 2 , which largely consists of…”
Effective Strategies in Zero-Shot Neural Machine Translation (2017, background): “State-of-the-art NMT systems have demonstrated that machine translation in many languages can achieve high quality results with large-scale data and sufﬁcient computational power[4, 5].”