The DOI matches its Crossref record; 14 of the 16 references it cites were found in a registry or an index.
Incorporating Uncertainty into Deep Learning for Spoken Language Assessment
Open accessAbstract
There is a growing demand for automatic assessment of spoken English proficiency.These systems need to handle large variations in input data owing to the wide range of candidate skill levels and L1s, and errors from ASR.Some candidates will be a poor match to the training data set, undermining the validity of the predicted grade.For high stakes tests it is essential for such systems not only to grade well, but also to provide a measure of their uncertainty in their predictions, enabling rejection to human graders.Previous work examined Gaussian Process (GP) graders which, though successful, do not scale well with large data sets.Deep Neural Networks (DNN) may also be used to provide uncertainty using Monte-Carlo Dropout (MCD).This paper proposes a novel method to yield uncertainty and compares it to GPs and DNNs with MCD.The proposed approach explicitly teaches a DNN to have low uncertainty on training data and high uncertainty on generated artificial data.On experiments conducted on data from the Business Language Testing Service (BULATS), the proposed approach is found to outperform GPs and DNNs with MCD in uncertainty-based rejection whilst achieving comparable grading performance.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“The area under this curve quantifies the overall quality of an UQ method (Malinin et al., 2017).”
From Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph (2025) · cites this paper as Malinin et al., 2017
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Adam: A Method for Stochastic Optimization
Kingma, D. P., & Ba, J. (2014) · arXiv · doi:10.48550/arxiv.1412.6980 · Cited by 98,946
Scikit-learn: Machine Learning in Python
Pedregosa, F., Varoquaux, G., Gramfort, A., Michel, V., Thirion, B., Grisel, O., Blondel, M., Müller, A., Nothman, J., Louppe, G., et al. (2012) · arXiv · doi:10.48550/arxiv.1201.0490 · Cited by 71,016 · cited 1×
“GPs were run using Scikit-Learn (Pedregosa et al., 2011) using a squared exponential covariance function.”
Method(2014) · Cited by 33,419
Auto-Encoding Variational Bayes
Kingma, D. P., & Welling, M. (2013) · arXiv · doi:10.48550/arxiv.1312.6114 · Cited by 20,067 · cited 1×
“However, the effect of the nature of artificial data on rejection performance should be further investigated and other data generation methods, such as Variational Auto-Encoders (Kingma and Welling, 2014), and metrics to assess similarity between artificial and real training data, examined.”
Method(2010) · Cited by 14,134
Gaussian Processes for Machine Learning
Rasmussen, C. E., & Williams, C. K. I. (2005) · doi:10.7551/mitpress/3206.001.0001 · Cited by 11,705
Dropout as a bayesian approximation: Representing model uncertainty in deep learning
Gal, Y., & Ghahramani, Z. (2015) · arXiv · doi:10.48550/arxiv.1506.02142 · Cited by 6,471 · cited 3×
“However, Gal and Ghahramani (2016) have shown that Monte-Carlo Dropout (MCD) can be used to derive an uncertainty estimate for a DNN. Alternatively, a Deep Density Network (DDN), which is a Mixture Density Network (Bishop, 1994) with only one mixture component, may be used to yield a mean and…”
BackgroundCommon European Framework of Reference for Languages: learning, teaching, assessment
Verhelst, N. D., Avermaet, P. V., Takala, S., Figueras, N., & North, B. (2009) · Ghent University Academic Bibliography (Ghent University) · Cited by 6,120
(1994) · Aston Publications Explorer (Aston University) · Cited by 1,245
Automatic scoring of non-native spontaneous speech in tests of spoken English
Zechner, K., Higgins, D., Xi, X., & Williamson, D. M. (2009) · Speech Communication · doi:10.1016/j.specom.2009.04.009 · Cited by 249 · cited 2×
“Previously, separate models were used to fil-ter out ”non-scorable” candidates (Yoon and Xie, 2014; Zechner et al., 2009; Higgins et al., 2011; Xie et al., 2012).”
Method
Research works printed but not linked: 2
- Diederik P. Kingma and Max Welling. 2014. Autoencoding variational bayes. In Proceedings of the 2nd International Conference on Learning Representations (ICLR) .
- Kevin P. Murphy. 2012. Machine Learning . The MIT Press.
Cited by
An automated essay scoring systems: a systematic literature review
Ramesh, D., & Sanampudi, S. K. (2022) · Artificial Intelligence Review · doi:10.1007/s10462-021-10068-2 · Cited by 550
Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution SamplesPreprint
Lee, K., Lee, H., Lee, K., & Shin, J. (2017) · arXiv (Cornell University) · doi:10.48550/arxiv.1711.09325 · Cited by 909
“n in our experiments, our method can be used for any classification tasks using deep neural networks. It is also interesting future directions applying our methods for other related tasks: regression (Malinin et al., 2017), network calibration (Guo et al., 2017), Bayesian probabilistic models (Li & Gal, 2017; Louiz…”
MethodPredictive Uncertainty Estimation via Prior Networks
Malinin, A., & Gales, M. (2018) · arXiv (Cornell University) · doi:10.48550/arxiv.1802.10501 · Cited by 337
“One solution is to synthetically generate points on the boundary of the in-domain region using a generative model [21, 22].”
BackgroundMethodInfluentialSelf-citationPredictive Uncertainty Estimation via Prior Networks
Malinin, A., & Gales, M. (2018) · Apollo (University of Cambridge) · doi:10.17863/cam.35237 · Cited by 163
Self-citationEnsemble Distribution DistillationPreprint
Gales, M., Malinin, A., & Mlodozeniec, B. (2019) · arXiv (Cornell University) · doi:10.17863/cam.49348 · Cited by 244
Self-citationA systematic data-driven Demand Side Management method for smart natural gas supply systems
Su, H., Zio, E., Zhang, J., Chi, L., Li, X., & Zhang, Z. (2019) · Energy Conversion and Management · doi:10.1016/j.enconman.2019.01.114 · Cited by 85
Uncertainty Estimation in Autoregressive Structured Prediction
Gales, M., & Malinin, A. (2021) · Apollo - University of Cambridge Repository · doi:10.17863/cam.63497 · Cited by 48
“1 PREDICTION REJECTION RATIO Here we describe the Prediction Rejection Ratio metric, proposed in (Malinin, 2019; Malinin et al., 2017), which in this work is used to assess how well measures of sequence-level uncertainty are able to identify sentences which are hard to translate/transcribe.”
MethodSelf-citationTowards automatic assessment of spontaneous spoken English
Wang, Y., Gales, M., Knill, K. M., Kyriakopoulos, K., Malinin, A., van Dalen, R. C., & Rashid, M. (2018) · Speech Communication · doi:10.1016/j.specom.2018.09.002 · Cited by 48
Self-citationGraph Posterior Network: Bayesian Predictive Uncertainty for Node ClassificationPreprint
Stadler, M., Charpentier, B., Geisler, S., Zügner, D., & Günnemann, S. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2110.14012 · Cited by 124
Benchmarking Uncertainty Quantification Methods for Large Language Models with LM-Polygraph
Vashurin, R., Fadeeva, E., Vazhentsev, A., Rvanova, L., Vasilev, D., Tsvigun, A., Petrakov, S., Xing, R., Sadallah, A., Grishchenkov, K., et al. (2025) · Transactions of the Association for Computational Linguistics · doi:10.1162/tacl_a_00737 · Cited by 21
“These pairs of uncertainty thresholds and associated average output quality give a prediction–rejection curve, and the area under this curve demonstrates the overall quality of a UQ method (Malinin et al., 2017).”
LM-Polygraph: Uncertainty estimation for language models
Fadeeva, E., Vashurin, R., Tsvigun, A., Vazhentsev, A., Petrakov, S., Fedyanin, K., Vasilev, D., Goncharova, E., Panchenko, A., Panov, M., et al. (2023) · Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing: System Demonstrations · doi:10.18653/v1/2023.emnlp-demo.41 · Cited by 20
Kumar, Y., Parekh, S., Singh, S., Li, J. J., Shah, R. R., & Chen, C. (2023) · Dialogue & Discourse · doi:10.5210/dad.2023.101 · Cited by 35
“However, while automatic scoring has seen much research in the recent years, model validation and testing still lag in the ML field with only a few contemporary works (Kumar et al., 2020; Pham et al., 2021; Yoon and Xie, 2014; Malinin et al., 2017).”
InfluentialSeq-UPS: Sequential Uncertainty-aware Pseudo-label Selection for Semi-Supervised Text Recognition
Patel, G., Allebach, J. P., & Qiu, Q. (2023) · IEEE Winter Conference on Applications of Computer Vision · doi:10.1109/wacv56688.2023.00612 · Cited by 30
“[41, 42, 43] proposed using the prediction rejection curves and the prediction rejection area ratios to qualitatively and quantitatively assess an uncertainty measure’s correlation to misclassifications, respectively.”
MethodGnowsis: Multimodal multitask learning for oral proficiency assessments
Takatsu, H., Suzuki, S., Eguchi, M., Matsuura, R., Saeki, M., & Matsuyama, Y. (2025) · Computer Speech & Language · doi:10.1016/j.csl.2025.101860 · Cited by 11
Universal Adversarial Attacks on Spoken Language Assessment Systems
Raina, V., Gales, M., & Knill, K. (2020) · Interspeech (USB) · doi:10.21437/interspeech.2020-1890 · Cited by 34
“Two feature-based graders were built; one GP-based [35] (GP txt ) and the other DNN-based [36] (DNN txt ).”
MethodSelf-citationEnsemble Distillation Approaches for Grammatical Error Correction
Fathullah, Y., Gales, M., & Malinin, A. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413385 · Cited by 12
Self-citationUncertainty in Structured Prediction.Preprint
Malinin, A., & Gales, M. (2020) · arXiv (Cornell University) · Cited by 7
Self-citationModelling Inter-Rater Uncertainty in Spoken Language Assessment
Wong, J. H. M., Zhang, H., & Chen, N. F. (2023) · IEEE/ACM Transactions on Audio Speech and Language Processing · doi:10.1109/taslp.2023.3297958 · Cited by 5
Bayesian deep learning: A model-based interpretable approach
Matsubara, T. (2019) · Nonlinear Theory and Its Applications IEICE · doi:10.1587/nolta.11.16 · Cited by 4
“The epistemic uncertainty is caused by an insufficient amount of data, and thus it appears far from data distribution and can detect out-of-distribution samples that are potentially mis-classified or with excessive errors [65, 66, 75].”
BackgroundOn the Calibration of Epistemic Uncertainty: Principles, Paradoxes and Conflictual Loss
Fellaji, M., Pennerath, F., Conan‐Guez, B., & Couceiro, M. (2024) · Lecture notes in computer science · doi:10.1007/978-3-031-70359-1_10 · Cited by 4
Analysing Bias in Spoken Language Assessment Using Concept Activation Vectors
Wei, X., Gales, M., & Knill, K. (2021) · IEEE International Conference on Acoustics Speech and Signal Processing · doi:10.1109/icassp39728.2021.9413988 · Cited by 8
“Significant progress has been made in applying various techniques from speech processing, machine learning (ML) and Artificial Intelligence (AI) to automate the language assessment process [1, 2, 3, 4].”
MethodSelf-citationAES Systems Are Both Overstable And Oversensitive: Explaining Why And Proposing DefensesPreprint
Singla, Y. K., Parekh, S., Singh, S., Li, J. J., Shah, R. R., & Chen, C. (2021) · arXiv (Cornell University) · doi:10.48550/arxiv.2109.11728 · Cited by 3
“However, while automatic scoring has seen much research in the recent years, model validation and testing still lag in the ML field with only a few contemporary works (Kumar et al., 2020; Pham et al., 2021; Yoon and Xie, 2014; Malinin et al., 2017).”
InfluentialSpatially-dependent Bayesian semantic perception under model and localization uncertainty
Feldman, Y., & Indelman, V. (2020) · Autonomous Robots · doi:10.1007/s10514-020-09921-0 · Cited by 2
“…used, among else, for efficient exploration and active learning Gal et al. (2017), Osband et al. (2016), safe decision making Lütjens et al. (2018), Malinin et al. (2017), McAllister et al. (2017), semantic segmentation Kendall et al. (2015a), camera localization Kendall et al. (2015b), and the…”
BackgroundImproved Auto-Marking Confidence for Spoken Language Assessment
Vecchio, M. D., Malinin, A., & Gales, M. (2018) · IEEE Spoken Language Technology Workshop (SLT) · doi:10.1109/slt.2018.8639634 · Cited by 1
“In this work, we adopt and extend the training procedure introduced in [12] as follows.”
BackgroundMethodSelf-citationUncertainty-Aware Pseudo-Labeling for Spoken Language Assessment
Lin, B., & Wang, L. (2021) · IEEE Workshop on Automatic Speech Recognition and Understanding (ASRU) · doi:10.1109/asru51503.2021.9687939
Holistic Uncertainty Estimation for Open-Set Recognition
Erlygin, L., & Zaytsev, A. A. (2026) · IEEE Access · doi:10.1109/access.2026.3654179
“A natural metric for evaluating the effectiveness of obtained quality scores is the area under a rejection curve [31], [32].”
The Cost of Reasoning: Chain-of-Thought Induces Overconfidence in Vision-Language Models
Welch, R., Konuk, E., & Smith, K. (2026) · Lecture notes in computer science · doi:10.1007/978-3-032-37196-6_27 · Cited by 5
Kobylyanskaya, S. (2024) · theses.fr (ABES) · doi:10.70675/5cefb8f5z36efz4263z8ef5za5ceb7b8f85c
Metadata
- Authors
- Malinin, A., Ragni, A., Knill, K., & Gales, M.
- Affiliations
- University of Cambridge (GB)
- Corresponding author
- Knill, Kate
- Venue
- Proceedings of the 55th Annual Meeting of the Association for Computational Linguistics (Volume 2: Short Papers)
- Published
- 2017
- Pages
- 45-50
- ISSN
- 0736-587X (ISSN-L)
- OpenAlex
- W2739575494
- Cited by
- 29
- References
- 16
- Citations by year
- 2026: 3 · 2025: 2 · 2024: 2 · 2023: 4 · 2021: 6 · 2020: 4 · 2019: 3 · 2018: 4 · 2017: 1
- FWCI
- 1.50
- Type
- conference paper
- Language
- en
- Open access
- hybrid
- doi.org
- Licence
- cc-by
- Locations
- Proceedings of the conference - Association for Computational Linguistics. Meeting (publishedVersion, cc-by)
- White Rose Research Online (University of Leeds, The University of Sheffield, University of York) (submittedVersion)
- Cambridge University Engineering Department Publications Database (submittedVersion)
- Apollo (University of Cambridge) (submittedVersion)
- Apollo
- Funders
- Engineering and Physical Sciences Research Council (1464018)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Speech Recognition and Synthesis
- Social Sciences › Arts and Humanities › Language and Linguistics › EFL/ESL Teaching and Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- uncertainty estimation, deep neural network, Monte Carlo dropout, Gaussian process regression, speech recognition errors, out-of-distribution data, high-stakes testing
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.