Language Model Transformers as Evaluators for Open-domain Dialogues

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Standard

Language Model Transformers as Evaluators for Open-domain Dialogues. / Nedelchev, Rostislav; Lehmann, Jens; Usbeck, Ricardo.
COLING 2020 - 28th International Conference on Computational Linguistics: Proceedings of the Conference. ed. / Donia Scott; Nuria Bel; Chengqing Zong. Association for Computational Linguistics (ACL), 2020. p. 6797-6808 (COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference).

Research output: Contributions to collected editions/worksArticle in conference proceedingsResearchpeer-review

Harvard

Nedelchev, R, Lehmann, J & Usbeck, R 2020, Language Model Transformers as Evaluators for Open-domain Dialogues. in D Scott, N Bel & C Zong (eds), COLING 2020 - 28th International Conference on Computational Linguistics: Proceedings of the Conference. COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference, Association for Computational Linguistics (ACL), pp. 6797-6808, 28th International Conference on Computational Linguistics, COLING 2020, Virtual, Online, Spain, 08.12.20. https://doi.org/10.18653/v1/2020.coling-main.599

APA

Nedelchev, R., Lehmann, J., & Usbeck, R. (2020). Language Model Transformers as Evaluators for Open-domain Dialogues. In D. Scott, N. Bel, & C. Zong (Eds.), COLING 2020 - 28th International Conference on Computational Linguistics: Proceedings of the Conference (pp. 6797-6808). (COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference). Association for Computational Linguistics (ACL). https://doi.org/10.18653/v1/2020.coling-main.599

Vancouver

Nedelchev R, Lehmann J, Usbeck R. Language Model Transformers as Evaluators for Open-domain Dialogues. In Scott D, Bel N, Zong C, editors, COLING 2020 - 28th International Conference on Computational Linguistics: Proceedings of the Conference. Association for Computational Linguistics (ACL). 2020. p. 6797-6808. (COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference). doi: 10.18653/v1/2020.coling-main.599

Bibtex

@inbook{060baa868fe74263b7f5495df8027644,
title = "Language Model Transformers as Evaluators for Open-domain Dialogues",
abstract = "Computer-based systems for communication with humans are a cornerstone of AI research since the 1950s. So far, the most effective way to assess the quality of the dialogues produced by these systems is to use resource-intensive manual labor instead of automated means. In this work, we investigate whether language models (LM) based on transformer neural networks can indicate the quality of a conversation. In a general sense, language models are methods that learn to predict one or more words based on an already given context. Due to their unsupervised nature, they are candidates for efficient, automatic indication of dialogue quality. We demonstrate that human evaluators have a positive correlation between the output of the language models and scores. We also provide some insights into their behavior and inner-working in a conversational context.",
keywords = "Informatics, Business informatics",
author = "Rostislav Nedelchev and Jens Lehmann and Ricardo Usbeck",
note = "We acknowledge the support of the EU projects Cleopatra (GA 812997) and TAILOR (GA 952215), the Federal Ministry for Economic Affairs and Energy (BMWi) project SPEAKER (FKZ 01MK20011A), the German Federal Ministry of Education and Research (BMBF) projects and excellence clusters ML2R (FKZ 01 15 18038 A/B/C), MLwin (01S18050 D/F), ScaDS.AI (01/S18026A) as well as the Fraunhofer Zukunftsstiftung project JOSEPH. Publisher Copyright: {\textcopyright} 2020 COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference. All rights reserved.; 28th International Conference on Computational Linguistics, COLING 2020 ; Conference date: 08-12-2020 Through 13-12-2020",
year = "2020",
month = jan,
day = "1",
doi = "10.18653/v1/2020.coling-main.599",
language = "English",
series = "COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference",
publisher = "Association for Computational Linguistics (ACL)",
pages = "6797--6808",
editor = "Donia Scott and Nuria Bel and Chengqing Zong",
booktitle = "COLING 2020 - 28th International Conference on Computational Linguistics",
address = "United States",
url = "https://coling2020.org, https://coling2020.org/COLING2020programme.pdf",

}

RIS

TY - CHAP

T1 - Language Model Transformers as Evaluators for Open-domain Dialogues

AU - Nedelchev, Rostislav

AU - Lehmann, Jens

AU - Usbeck, Ricardo

N1 - We acknowledge the support of the EU projects Cleopatra (GA 812997) and TAILOR (GA 952215), the Federal Ministry for Economic Affairs and Energy (BMWi) project SPEAKER (FKZ 01MK20011A), the German Federal Ministry of Education and Research (BMBF) projects and excellence clusters ML2R (FKZ 01 15 18038 A/B/C), MLwin (01S18050 D/F), ScaDS.AI (01/S18026A) as well as the Fraunhofer Zukunftsstiftung project JOSEPH. Publisher Copyright: © 2020 COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference. All rights reserved.

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Computer-based systems for communication with humans are a cornerstone of AI research since the 1950s. So far, the most effective way to assess the quality of the dialogues produced by these systems is to use resource-intensive manual labor instead of automated means. In this work, we investigate whether language models (LM) based on transformer neural networks can indicate the quality of a conversation. In a general sense, language models are methods that learn to predict one or more words based on an already given context. Due to their unsupervised nature, they are candidates for efficient, automatic indication of dialogue quality. We demonstrate that human evaluators have a positive correlation between the output of the language models and scores. We also provide some insights into their behavior and inner-working in a conversational context.

AB - Computer-based systems for communication with humans are a cornerstone of AI research since the 1950s. So far, the most effective way to assess the quality of the dialogues produced by these systems is to use resource-intensive manual labor instead of automated means. In this work, we investigate whether language models (LM) based on transformer neural networks can indicate the quality of a conversation. In a general sense, language models are methods that learn to predict one or more words based on an already given context. Due to their unsupervised nature, they are candidates for efficient, automatic indication of dialogue quality. We demonstrate that human evaluators have a positive correlation between the output of the language models and scores. We also provide some insights into their behavior and inner-working in a conversational context.

KW - Informatics

KW - Business informatics

UR - http://www.scopus.com/inward/record.url?scp=85108285068&partnerID=8YFLogxK

UR - https://www.mendeley.com/catalogue/0f9694bb-370d-3c37-bb25-8347d9aac64a/

U2 - 10.18653/v1/2020.coling-main.599

DO - 10.18653/v1/2020.coling-main.599

M3 - Article in conference proceedings

AN - SCOPUS:85108285068

T3 - COLING 2020 - 28th International Conference on Computational Linguistics, Proceedings of the Conference

SP - 6797

EP - 6808

BT - COLING 2020 - 28th International Conference on Computational Linguistics

A2 - Scott, Donia

A2 - Bel, Nuria

A2 - Zong, Chengqing

PB - Association for Computational Linguistics (ACL)

T2 - 28th International Conference on Computational Linguistics, COLING 2020

Y2 - 8 December 2020 through 13 December 2020

ER -

Recently viewed

Publications

  1. Integrierte Eingabegeräte
  2. Personal need for structure as a boundary condition for humor in leadership
  3. Careless product use in access-based services
  4. Separating Cognitive and Content Domains in Mathematical Competence
  5. Overview of a Proposed Ecological Risk Assessment Process for Honey bees (Apis mellifera) and Non‐Apis Bees
  6. Toward spatial fit in the governance of global commodity flows
  7. How young children integrate information sources to infer the meaning of words
  8. Method of Artificial Vision in Guide Cane for Visually Impaired People
  9. Precipitation Kinetics of AA6082: An Experimental and Numerical Investigation
  10. Calibration of a simple method for determining ammonia volatilization in the field - Comparative measurements in Henan Province, China
  11. Friction riveting of 3D printed polyamide 6 with AA 6056-T6
  12. New methods for the analysis of links between international firm activities and firm performance
  13. A transfer operator based numerical investigation of coherent structures in three-dimensional Southern ocean circulation
  14. Optimization and Validation of an LC Method for the Determination of Cefdinir in Dosage Form and Human Urine
  15. Zur Methodologie der ‘Fehleranalyse’ in der mathematikdidaktischen Forschung
  16. Case study on delivery time determination using a machine learning approach in small batch production companies
  17. A Robust Decoupling Estimator to Indentify Electrical Parameters for Three-Phase Permanent Magnet Synchronous Motors
  18. Self-regulated learning and self assessment in online mathematics bridging courses
  19. Anticipated imitation of multiple agents
  20. Increasing personal initiative in small business managers or owners leads to entrepreneurial success: A theory-based controlled randomized field intervention for evidence-based management
  21. Effects of tree diversity on canopy space occupation vary with tree size and canopy space definition in a mature broad-leaved forest
  22. Environmental performance, carbon performance and earnings management
  23. Dimension theoretical properties of generalized Baker's transformations
  24. Cascade MIMO P-PID Controllers Applied in an Over-actuated Quadrotor Tilt-Rotor
  25. Trade Dynamics, Trade Costs and Market Size: First Evidence from the Exporter and Importer Dynamics Database for Germany
  26. Motion-decoupled internal force control in grasping with visco-elastic contacts
  27. Nitrogen uptake by grassland communities
  28. Deliberative mapping of ecosystem services within and around Doñana National Park (SW Spain) in relation to land use change
  29. The role of the situation model in mathematical modelling
  30. Investigation of the deformation behavior of Fe-3%Si sheet metal with large grains via crystal plasticity and finite-element modeling
  31. Effects of Soil Properties, Temperature and Disturbance on Diversity and Functional Composition of Plant Communities Along a Steep Elevational Gradient on Tenerife
  32. PyFin-sentiment
  33. Green deserts, but not always
  34. Evidence on copula-based double-hurdle models with flexible margins
  35. Deep Rolling for Tailoring Residual Stresses of AA2024 Sheet Metals
  36. User experience predicts the effectiveness of a gamified recovery app
  37. A PD Fuzzy Control of a Nonholonomic Car-Like Robot for Drive Assistant Systems
  38. Land-use change differentially affects endemic, forest andopen-land butterflies in Madagascar
  39. Modeling of Friction-Induced Vibrations during Tightening of Bolted Joints