Confidence Measures for Interactive Neural Machine Translation

Navarro, Angel F.; Casacuberta, Francisco

doi:10.21437/iberspeech.2021-42

Cited by 4 publications

(15 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Sanchis-Trilles et al have only proved the validity of the technique in conjunction with Statistical Machine Translation (SMT) models in their study. More recently, [18] have tested MAs with NMT models at the word level.…”

Section: Iter-0 (P) (ŝ H ) ( )mentioning

confidence: 99%

“…This work achieved a significant reduction in human effort. More recently, [18] have implemented this technique in an IPNMT framework and conducted a comparison of the results, obtaining a higher reduction in human effort. With advances in technology, NMT models generate higher quality translations, thus improving the probability that the system can fix errors without any extra information.…”

Section: Related Workmentioning

confidence: 99%

“…To quantify the effort required by the user when typing the correct characters, we used the Character Stroke Ratio (CSR) metric. In Section 5.5, we will compare our CSR reduction with the Word Stroke Ratio (WSR) reduction reported in a previous work [18], as both represent the human effort performed. We also used metrics to account for the mouse actions, including the character Mouse Action Ratio (cMAR) and the useful Mouse Action Ratio (uMAR).…”

Section: System Evaluationmentioning

confidence: 99%

“…At present, with the advent of neural models, research on such workbenches seems to have stopped; however, research on new techniques to reduce human effort in the IPNMT field is still being carried out. Some projects have changed the displays used in the translation sessions to make the process easier and more intuitive [9,10], while others generate additional data that facilitate the translation process [11,12]. Finally, other projects have studied new and minimalist feedback formats to provide to the machine [13,14].…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

On the Use of Mouse Actions at the Character Level

Navarro

Casacuberta

2022

Information

Self Cite

View full text Add to dashboard Cite

Neural Machine Translation (NMT) has improved performance in several tasks up to human parity. However, many companies still use Computer-Assisted Translation (CAT) tools to achieve perfect translation, as well as other tools. Among these tools, we find Interactive-Predictive Neural Machine Translation (IPNMT) systems, whose main feature is facilitating machine–human interactions. In the most conventional systems, the human user fixes a translation error by typing the correct word, sending this feedback to the machine which generates a new translation that satisfies it. In this article, we remove the necessity of typing to correct translations by using the bandit feedback obtained from the cursor position when the user performs a Mouse Action (MA). Our system generates a new translation that fixes the error using only the error position. The user can perform multiple MAs at the same position if the error is not fixed, each of which increases the correction probability. One of the main objectives in the IPNMT field is reducing the required human effort, in order to optimize the translation time. With the proposed technique, an 84% reduction in the number of keystrokes performed can be achieved, while still generating perfect translations. For this reason, we recommend the use of this technique in IPNMT systems.

show abstract

Section: Iter-0 (P) (ŝ H ) ( )mentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

Section: System Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the Use of Mouse Actions at the Character Level

Navarro

Casacuberta

2022

Information

Self Cite

View full text Add to dashboard Cite

show abstract

“…Kreutzer and Riezler (2019) proposed the use of self-regulation strategies that learn which type of feedback to query from a human teacher. The use of confidence measures ( CM) was integrated into interactive neural machine translation ( INMT) (Knowles and Koehn, 2018;Navarro and Casacuberta, 2021a). Santy et al (2019) presented a demonstrator of an INMT system that assists human translators with on-the-fly hints and suggestions.…”

Section: Prefix-based Imtmentioning

confidence: 99%

Some Contributions to Interactive Machine Translation and to the Applications of Machine Translation for Historical Documents

Ballester¹

View full text Add to dashboard Cite

The research leading to this thesis has been partially funded by Ministerio de Economía y Competitividad (MINECO) under projects SmartWays (grant agreement RTC-2014-1466-4), CoMUN-HaT (grant agreement TIN2015-70924-C2-1-R) and MISMISFAKEnHATE (grant agreement PGC2018-096212-B-C31); Generalitat Valenciana under projects ALMAMATER (grant agreement PROME-TEOII/2014/030) and DeepPattern (grant agreement PROMETEO/2019/121); the European Union through Programa Operativo del Fondo Europeo de Desarrollo Regional (FEDER) from Comunitat Valenciana (2014-2020) under project Sistemas de frabricación inteligentes para la indústria 4.0 (grant agreement ID-IFEDER/2018/025); and the PRHLT research center under the research line Machine Learning Applications. We gratefully acknowledge the support of NVIDIA Corporation with the donation of a GPU used through this thesis. Additionally, I would like to thank Andrés Trapiello and Ediciones Destino for granting us permission to use their book in our researches.

show abstract

Turning Machine Translation Metrics into Confidence Measures

Martínez

Nolla

2022

Natural Language Processing and Information Systems

View full text Add to dashboard Cite

Confidence Measures for Interactive Neural Machine Translation

Cited by 4 publications

References 0 publications

On the Use of Mouse Actions at the Character Level

On the Use of Mouse Actions at the Character Level

Some Contributions to Interactive Machine Translation and to the Applications of Machine Translation for Historical Documents

Turning Machine Translation Metrics into Confidence Measures

Contact Info

Product

Resources

About