Zobrazeno 1 - 10
of 17
pro vyhledávání: '"Wuebker, Joern"'
Reinforcement Learning from Human Feedback (RLHF) and derivative techniques like Direct Preference Optimization (DPO) are task-alignment algorithms used to repurpose general, foundational models for specific tasks. We show that applying task-alignmen
Externí odkaz:
http://arxiv.org/abs/2409.17673
The emergent ability of Large Language Models to use a small number of examples to learn to perform in novel domains and tasks, also called in-context learning (ICL). In this work, we show that a much smaller model can be trained to perform ICL by fi
Externí odkaz:
http://arxiv.org/abs/2309.08590
We introduce translation error correction (TEC), the task of automatically correcting human-generated translations. Imperfections in machine translations (MT) have long motivated systems for improving translations post-hoc with automatic post-editing
Externí odkaz:
http://arxiv.org/abs/2206.08593
Autor:
Läubli, Samuel, Simianer, Patrick, Wuebker, Joern, Kovacs, Geza, Sennrich, Rico, Green, Spence
Widely used computer-aided translation (CAT) tools divide documents into segments such as sentences and arrange them in a side-by-side, spreadsheet-like view. We present the first controlled evaluation of these design choices on translator performanc
Externí odkaz:
http://arxiv.org/abs/2011.05978
Word alignment was once a core unsupervised learning task in natural language processing because of its essential role in training statistical machine translation (MT) models. Although unnecessary for training neural MT models, word alignment still p
Externí odkaz:
http://arxiv.org/abs/2004.14675
Multi-layer models with multiple attention heads per layer provide superior translation quality compared to simpler and shallower models, but determining what source context is most relevant to each target word is more challenging as a result. Theref
Externí odkaz:
http://arxiv.org/abs/1901.11359
This work systematically analyzes the smoothing effect of vocabulary reduction for phrase translation models. We extensively compare various word-level vocabularies to show that the performance of smoothing is not significantly affected by the choice
Externí odkaz:
http://arxiv.org/abs/1901.01574
We propose and compare methods for gradient-based domain adaptation of self-attentive neural machine translation models. We demonstrate that a large proportion of model parameters can be frozen during adaptation with minimal or no reduction in transl
Externí odkaz:
http://arxiv.org/abs/1811.01990
Akademický článek
Tento výsledek nelze pro nepřihlášené uživatele zobrazit.
K zobrazení výsledku je třeba se přihlásit.
K zobrazení výsledku je třeba se přihlásit.
Autor:
Läubli, Samuel, Simianer, Patrick, Wuebker, Joern, Kovacs, Geza, Sennrich, Rico, Green, Spence
Publikováno v:
Target: International Journal on Translation Studies; 2022, Vol. 34 Issue 2, p309-342, 34p