1 code implementation • 29 Jan 2024 • Nikita Moghe, Arnisa Fazla, Chantal Amrhein, Tom Kocmi, Mark Steedman, Alexandra Birch, Rico Sennrich, Liane Guillou
We benchmark metric performance, assess their incremental performance over successive campaigns, and measure their sensitivity to a range of linguistic phenomena.
1 code implementation • 28 Nov 2023 • Noëmi Aepli, Chantal Amrhein, Florian Schottmann, Rico Sennrich
For sensible progress in natural language processing, it is important that we are aware of the limitations of the evaluation metrics we use.
no code implementations • 2 Nov 2023 • Chantal Amrhein, Nikita Moghe, Liane Guillou
We benchmark the performance of segmentlevel metrics submitted to WMT 2023 using the ACES Challenge Set (Amrhein et al., 2022).
1 code implementation • 6 Jun 2023 • Janis Goldzycher, Moritz Preisig, Chantal Amrhein, Gerold Schneider
In this paper, we test whether natural language inference (NLI) models which perform well in zero- and few-shot settings can benefit hate speech detection performance in scenarios where only a limited amount of labeled data is available in the target language.
1 code implementation • 18 May 2023 • Chantal Amrhein, Florian Schottmann, Rico Sennrich, Samuel Läubli
We hypothesise that creating training data in the reverse direction, i. e. starting from gender-fair text, is easier for morphologically complex languages and show that it matches the performance of state-of-the-art rewriting models for English.
1 code implementation • 27 Oct 2022 • Chantal Amrhein, Nikita Moghe, Liane Guillou
As machine translation (MT) metrics improve their correlation with human judgement every year, it is crucial to understand the limitations of such metrics at the segment level.
Ranked #1 on Machine Translation on ACES
1 code implementation • 24 Oct 2022 • Chantal Amrhein, Barry Haddow
For real-life applications, it is crucial that end-to-end spoken language translation models perform well on continuous audio, without relying on human-supplied segmentation.
1 code implementation • 10 Feb 2022 • Chantal Amrhein, Rico Sennrich
Neural metrics have achieved impressive correlation with human judgements in the evaluation of machine translation systems, but before we can safely optimise towards such metrics, we should be aware of (and ideally eliminate) biases toward bad translations that receive high scores.
1 code implementation • Findings (EMNLP) 2021 • Chantal Amrhein, Rico Sennrich
Data-driven subword segmentation has become the default strategy for open-vocabulary machine translation and other NLP tasks, but may not be sufficiently generic for optimal learning of non-concatenative morphology.
1 code implementation • NAACL 2021 • Annette Rios, Chantal Amrhein, Noëmi Aepli, Rico Sennrich
Many sequence-to-sequence tasks in natural language processing are roughly monotonic in the alignment between source and target sequence, and previous work has facilitated or enforced learning of monotonic attention behavior via specialized attention functions or pretraining.
no code implementations • Findings of the Association for Computational Linguistics 2020 • Chantal Amrhein, Rico Sennrich
Our results show that romanization entails information loss and is thus not always superior to simpler vocabulary transfer methods, but can improve the transfer between related languages with different scripts.
no code implementations • WS 2019 • Samuel Läubli, Chantal Amrhein, Patrick Düggelin, Beatriz Gonzalez, Alena Zwahlen, Martin Volk
Neural machine translation (NMT) has set new quality standards in automatic translation, yet its effect on post-editing productivity is still pending thorough investigation.