Search Results for author: Julia Kreutzer

Found 42 papers, 20 papers with code

Bandits Don’t Follow Rules: Balancing Multi-Facet Machine Translation with Multi-Armed Bandits

no code implementations Findings (EMNLP) 2021 Julia Kreutzer, David Vilar, Artem Sokolov

Training data for machine translation (MT) is often sourced from a multitude of large corpora that are multi-faceted in nature, e. g. containing contents from multiple domains or different levels of quality or complexity.

Machine Translation Multi-Armed Bandits +1

RLHF Can Speak Many Languages: Unlocking Multilingual Preference Optimization for LLMs

no code implementations2 Jul 2024 John Dang, Arash Ahmadian, Kelly Marchisio, Julia Kreutzer, Ahmet Üstün, Sara Hooker

Preference optimization techniques have become a standard final stage for training state-of-art large language models (LLMs).

Cross-Lingual Transfer

LLM See, LLM Do: Guiding Data Generation to Target Non-Differentiable Objectives

1 code implementation1 Jul 2024 Luísa Shimabucoro, Sebastian Ruder, Julia Kreutzer, Marzieh Fadaee, Sara Hooker

Our findings invite the question can we explicitly steer the models towards the properties we want at test time by exploiting the data generation process?

Data Integration

The Multilingual Alignment Prism: Aligning Global and Local Preferences to Reduce Harm

no code implementations26 Jun 2024 Aakanksha, Arash Ahmadian, Beyza Ermis, Seraphina Goldfarb-Tarrant, Julia Kreutzer, Marzieh Fadaee, Sara Hooker

We collect the first set of human annotated red-teaming prompts in different languages distinguishing between global and local harm, which serve as a laboratory for understanding the reliability of alignment techniques when faced with preference distributions that are non-stationary across geographies and languages.

Cross-Lingual Transfer

Back to Basics: Revisiting REINFORCE Style Optimization for Learning from Human Feedback in LLMs

no code implementations22 Feb 2024 Arash Ahmadian, Chris Cremer, Matthias Gallé, Marzieh Fadaee, Julia Kreutzer, Olivier Pietquin, Ahmet Üstün, Sara Hooker

AI alignment in the shape of Reinforcement Learning from Human Feedback (RLHF) is increasingly treated as a crucial ingredient for high performance large language models.

JoeyS2T: Minimalistic Speech-to-Text Modeling with JoeyNMT

2 code implementations5 Oct 2022 Mayumi Ohta, Julia Kreutzer, Stefan Riezler

JoeyS2T is a JoeyNMT extension for speech-to-text tasks such as automatic speech recognition and end-to-end speech translation.

Automatic Speech Recognition Automatic Speech Recognition (ASR) +5

Can Multilinguality benefit Non-autoregressive Machine Translation?

no code implementations16 Dec 2021 Sweta Agrawal, Julia Kreutzer, Colin Cherry

Non-autoregressive (NAR) machine translation has recently achieved significant improvements, and now outperforms autoregressive (AR) models on some benchmarks, providing an efficient alternative to AR inference.

Machine Translation Translation

Bandits Don't Follow Rules: Balancing Multi-Facet Machine Translation with Multi-Armed Bandits

no code implementations13 Oct 2021 Julia Kreutzer, David Vilar, Artem Sokolov

Training data for machine translation (MT) is often sourced from a multitude of large corpora that are multi-faceted in nature, e. g. containing contents from multiple domains or different levels of quality or complexity.

Machine Translation Multi-Armed Bandits +1

Modelling Latent Translations for Cross-Lingual Transfer

1 code implementation23 Jul 2021 Edoardo Maria Ponti, Julia Kreutzer, Ivan Vulić, Siva Reddy

To remedy this, we propose a new technique that integrates both steps of the traditional pipeline (translation and classification) into a single model, by treating the intermediate translations as a latent random variable.

Cross-Lingual Transfer Few-Shot Learning +5

Revisiting the Weaknesses of Reinforcement Learning for Neural Machine Translation

1 code implementation NAACL 2021 Samuel Kiegeland, Julia Kreutzer

Policy gradient algorithms have found wide adoption in NLP, but have recently become subject to criticism, doubting their suitability for NMT.

Domain Adaptation Machine Translation +4

MasakhaNER: Named Entity Recognition for African Languages

2 code implementations22 Mar 2021 David Ifeoluwa Adelani, Jade Abbott, Graham Neubig, Daniel D'souza, Julia Kreutzer, Constantine Lignos, Chester Palen-Michel, Happy Buzaaba, Shruti Rijhwani, Sebastian Ruder, Stephen Mayhew, Israel Abebe Azime, Shamsuddeen Muhammad, Chris Chinenye Emezue, Joyce Nakatumba-Nabende, Perez Ogayo, Anuoluwapo Aremu, Catherine Gitau, Derguene Mbaye, Jesujoba Alabi, Seid Muhie Yimam, Tajuddeen Gwadabe, Ignatius Ezeani, Rubungo Andre Niyongabo, Jonathan Mukiibi, Verrah Otiende, Iroro Orife, Davis David, Samba Ngom, Tosin Adewumi, Paul Rayson, Mofetoluwa Adeyemi, Gerald Muriuki, Emmanuel Anebi, Chiamaka Chukwuneke, Nkiruka Odu, Eric Peter Wairagala, Samuel Oyerinde, Clemencia Siro, Tobius Saul Bateesa, Temilola Oloyede, Yvonne Wambui, Victor Akinode, Deborah Nabagereka, Maurice Katusiime, Ayodele Awokoya, Mouhamadane MBOUP, Dibora Gebreyohannes, Henok Tilaye, Kelechi Nwaike, Degaga Wolde, Abdoulaye Faye, Blessing Sibanda, Orevaoghene Ahia, Bonaventure F. P. Dossou, Kelechi Ogueji, Thierno Ibrahima DIOP, Abdoulaye Diallo, Adewale Akinfaderin, Tendai Marengereke, Salomey Osei

We take a step towards addressing the under-representation of the African continent in NLP research by creating the first large publicly available high-quality dataset for named entity recognition (NER) in ten African languages, bringing together a variety of stakeholders.

named-entity-recognition Named Entity Recognition +2

Inference Strategies for Machine Translation with Conditional Masking

no code implementations EMNLP 2020 Julia Kreutzer, George Foster, Colin Cherry

Conditional masked language model (CMLM) training has proven successful for non-autoregressive and semi-autoregressive sequence generation tasks, such as machine translation.

Language Modelling Machine Translation +1

On Optimal Transformer Depth for Low-Resource Language Translation

1 code implementation9 Apr 2020 Elan van Biljon, Arnu Pretorius, Julia Kreutzer

Therefore, by showing that transformer models perform well (and often best) at low-to-moderate depth, we hope to convince fellow researchers to devote less computational resources, as well as time, to exploring overly large models during the development of these systems.

Low Resource NMT NMT +1

Joey NMT: A Minimalist NMT Toolkit for Novices

8 code implementations IJCNLP 2019 Julia Kreutzer, Jasmijn Bastings, Stefan Riezler

We present Joey NMT, a minimalist neural machine translation toolkit based on PyTorch that is specifically designed for novices.

General Knowledge Machine Translation +2

Self-Regulated Interactive Sequence-to-Sequence Learning

7 code implementations ACL 2019 Julia Kreutzer, Stefan Riezler

Not all types of supervision signals are created equal: Different types of feedback have different costs and effects on learning.

Active Learning Machine Translation +1

Explaining and Generalizing Back-Translation through Wake-Sleep

no code implementations12 Jun 2018 Ryan Cotterell, Julia Kreutzer

Back-translation has become a commonly employed heuristic for semi-supervised neural machine translation.

Machine Translation Translation

Reliability and Learnability of Human Bandit Feedback for Sequence-to-Sequence Reinforcement Learning

1 code implementation ACL 2018 Julia Kreutzer, Joshua Uyheng, Stefan Riezler

We present a study on reinforcement learning (RL) from human bandit feedback for sequence-to-sequence learning, exemplified by the task of bandit neural machine translation (NMT).

Machine Translation NMT +3

A Reinforcement Learning Approach to Interactive-Predictive Neural Machine Translation

1 code implementation3 May 2018 Tsz Kin Lam, Julia Kreutzer, Stefan Riezler

We present an approach to interactive-predictive neural machine translation that attempts to reduce human effort from three directions: Firstly, instead of requiring humans to select, correct, or delete segments, we employ the idea of learning from human reinforcements in form of judgments on the quality of partial translations.

Machine Translation reinforcement-learning +2

Can Neural Machine Translation be Improved with User Feedback?

no code implementations NAACL 2018 Julia Kreutzer, Shahram Khadivi, Evgeny Matusov, Stefan Riezler

We present the first real-world application of methods for improving neural machine translation (NMT) with human reinforcement, based on explicit and implicit user feedback collected on the eBay e-commerce platform.

Machine Translation NMT +1

Bandit Structured Prediction for Neural Sequence-to-Sequence Learning

1 code implementation ACL 2017 Julia Kreutzer, Artem Sokolov, Stefan Riezler

Bandit structured prediction describes a stochastic optimization framework where learning is performed from partial feedback.

Domain Adaptation Machine Translation +3

Stochastic Structured Prediction under Bandit Feedback

1 code implementation NeurIPS 2016 Artem Sokolov, Julia Kreutzer, Christopher Lo, Stefan Riezler

Stochastic structured prediction under bandit feedback follows a learning protocol where on each of a sequence of iterations, the learner receives an input, predicts an output structure, and receives partial feedback in form of a task loss evaluation of the predicted structure.

Structured Prediction

Cannot find the paper you are looking for? You can Submit a new open access paper.