no code implementations • ICON 2019 • Niyati Bafna, Dipti Sharma
English-Hindi machine translation systems have difficulty interpreting verb phrase ellipsis (VPE) in English, and commit errors in translating sentences with VPE.
no code implementations • NAACL (CMCL) 2021 • Kartik Sharma, Niyati Bafna, Samar Husain
The models differ in their use of prior context during the prediction process – the context is either noisy or noise-free.
no code implementations • ICON 2021 • Niyati Bafna, Martin Vastl, Ondřej Bojar
Technical terms may require special handling when the target audience is bilingual, depending on the cultural and educational norms of the society in question.
no code implementations • LREC 2022 • Zdeněk Žabokrtský, Niyati Bafna, Jan Bodnár, Lukáš Kyjánek, Emil Svoboda, Magda Ševčíková, Jonáš Vidra
Our work aims at developing a multilingual data resource for morphological segmentation.
no code implementations • 16 Mar 2024 • Niyati Bafna, Philipp Koehn, David Yarowsky
While Transformer-based neural machine translation (NMT) is very effective in high-resource settings, many languages lack the necessary large parallel corpora to benefit from it.
1 code implementation • 23 May 2023 • Niyati Bafna, Cristina España-Bonet, Josef van Genabith, Benoît Sagot, Rachel Bawden
Most existing approaches for unsupervised bilingual lexicon induction (BLI) depend on good quality static or contextual embeddings requiring large monolingual corpora for both languages.