Iryna Gurevych
Technische Universität Darmstadt
Machine learningSemantic similarityWorld Wide WebGermanArtificial intelligenceDomain (software engineering)Natural language processingInformation retrievalData scienceAnnotationSentenceQuality (business)Context (language use)Task (project management)ArgumentComputer scienceLinguisticsQuestion answeringWord (computer architecture)Argumentation theory
566Publications
66H-index
10.7kCitations
Publications 505
Newest
#1Michael Bugert (Technische Universität Darmstadt)H-Index: 4
#2Nils ReimersH-Index: 16
Last. Iryna GurevychH-Index: 66
view all 3 authors...
Cross-document event coreference resolution (CDCR) is an NLP task in which mentions of events need to be identified and clustered throughout a collection of documents. CDCR aims to benefit downstream multi-document applications, but despite recent progress on corpora and system development, downstream improvements from applying CDCR have not been shown yet. We make the observation that every CDCR system to date was developed, trained, and tested only on a single respective corpus. This raises st...
Source
Nov 1, 2021 in EMNLP (Empirical Methods in Natural Language Processing)
#1Prasetya Ajie Utama (Technische Universität Darmstadt)H-Index: 7
#2Nafise Sadat Moosavi (Technische Universität Darmstadt)H-Index: 8
Last. Iryna Gurevych (University of Paderborn)H-Index: 66
view all 4 authors...
Recent prompt-based approaches allow pretrained language models to achieve strong performances on few-shot finetuning by reformulating downstream tasks as a language modeling problem. In this work, we demonstrate that, despite its advantages on low data regimes, finetuned prompt-based models for sentence pair classification tasks still suffer from a common pitfall of adopting inference heuristics based on lexical overlap, e.g., models incorrectly assuming a sentence pair is of the same meaning b...
Reasoning over multiple modalities, e.g. in Visual Question Answering (VQA), requires an alignment of semantic concepts across domains. Despite the widespread success of end-to-end learning, today's multimodal pipelines by and large leverage pre-extracted, fixed features from object detectors, typically Faster R-CNN, as representations of the visual world. The obvious downside is that the visual representation is not specifically tuned to the multimodal task at hand. At the same time, while tran...
#1Kevin StoweH-Index: 8
Last. Iryna GurevychH-Index: 66
view all 3 authors...
#1Jonas PfeifferH-Index: 11
#2Gregor GeigleH-Index: 2
Last. Iryna GurevychH-Index: 66
view all 7 authors...
Recent advances in multimodal vision and language modeling have predominantly focused on the English language, mostly due to the lack of multilingual multimodal datasets to steer modeling efforts. In this work, we address this gap and provide xGQA, a new multilingual evaluation benchmark for the visual question answering task. We extend the established English GQA dataset to 7 typologically diverse languages, enabling us to detect and explore crucial challenges in cross-lingual visual question a...
#1Mohsen MesgarH-Index: 7
Last. Iryna GurevychH-Index: 66
view all 3 authors...
Sep 9, 2021 in EMNLP (Empirical Methods in Natural Language Processing)
#1Leonardo F. R. Ribeiro (Technische Universität Darmstadt)H-Index: 10
#2Jonas Pfeiffer (Technische Universität Darmstadt)H-Index: 11
Last. Iryna Gurevych (University of Paderborn)H-Index: 66
view all 4 authors...
Recent work on multilingual AMR-to-text generation has exclusively focused on data augmentation strategies that utilize silver AMR. However, this assumes a high quality of generated AMRs, potentially limiting the transferability to the target task. In this paper, we investigate different techniques for automatically generating AMR annotations, where we aim to study which source of information yields better multilingual results. Our models trained on gold AMR with silver (machine translated) sent...
#2Edwin Simpson (UoB: University of Bristol)H-Index: 12
Peer review is the main quality control mechanism in academia. Quality of scientific work has many dimensions; coupled with the subjective nature of the reviewing task, this makes final decision making based on the reviews and scores therein very difficult and time-consuming. To assist with this important task, we cast it as a paper ranking problem based on peer review texts and reviewer scores. We introduce a novel, multi-faceted generic evaluation framework for making final decisions based on ...
Aug 26, 2021 in EMNLP (Empirical Methods in Natural Language Processing)
#1Andreas Rücklé (Technische Universität Darmstadt)H-Index: 11
#2Gregor GeigleH-Index: 2
Last. Iryna Gurevych (University of Paderborn)H-Index: 66
view all 7 authors...
Transformer models are expensive to fine-tune, slow for inference, and have large storage requirements. Recent approaches tackle these shortcomings by training smaller models, dynamically reducing the model size, and by training light-weight adapters. In this paper, we propose AdapterDrop, removing adapters from lower transformer layers during training and inference, which incorporates concepts from all three directions. We show that AdapterDrop can dynamically reduce the computational overhead ...
Aug 26, 2021 in EMNLP (Empirical Methods in Natural Language Processing)
#1Jonas Pfeiffer (Technische Universität Darmstadt)H-Index: 11
#2Ivan Vulić (University of Cambridge)H-Index: 33
Last. Sebastian Ruder (Google)H-Index: 34
view all 4 authors...
Massively multilingual language models such as multilingual BERT offer state-of-the-art cross-lingual transfer performance on a range of NLP tasks. However, due to limited capacity and large differences in pretraining data sizes, there is a profound performance gap between resource-rich and resource-poor target languages. The ultimate challenge is dealing with under-resourced languages not covered at all by the models and written in scripts unseen during pretraining. In this work, we propose a s...
This website uses cookies.
We use cookies to improve your online experience. By continuing to use our website we assume you agree to the placement of these cookies.
To learn more, you can find in our Privacy Policy.