The DOI matches its Crossref record; 83 of the 83 references it cites were found in a registry or an index.
Explaining machine learning models with interactive natural language conversations using TalkToModel
Open accessPublished version of a preprint
Abstract
Practitioners increasingly use machine learning (ML) models, yet models have become more complex and harder to understand. To understand complex models, researchers have proposed techniques to explain model predictions. However, practitioners struggle to use explainability methods because they do not know which explanation to choose and how to interpret the explanation. Here we address the challenge of using explainability methods by proposing TalkToModel: an interactive dialogue system that explains ML models through natural language conversations. TalkToModel consists of three components: an adaptive dialogue engine that interprets natural language and generates meaningful responses; an execution component that constructs the explanations used in the conversation; and a conversational interface. In real-world evaluations, 73% of healthcare workers agreed they would use TalkToModel over existing systems for understanding a disease prediction model, and 85% of ML professionals agreed TalkToModel was easier to use, demonstrating that TalkToModel is highly effective for model explainability.
Verifying this paper
8 of 9 steps done · this page updates itself — you can leave and come back
- Registry recordVerified
- Full textWe couldn’t retrieve the full text
- Reference list80 references found
- Reference checkChecking each cited work — 83 of 83
- StatisticsNeeds the full text
- ClaimsNeeds the full text
- Cited sourcesNeeds claims to check
- Self-checkNeeds the full text
- VerificationFindings verified
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
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 few studies that move beyond purely local explanations [11, 74,80,81,104,116,125,129] point to promising but underexplored directions.”
From Explainability of decoder-only clinical large language models: A scoping review (2026) · cites this paper as [104]
In the citing paper
“Of the 69 reviewed studies, only three [75,76,104] included any user-facing evaluation of explanation presentation.”
From Explainability of decoder-only clinical large language models: A scoping review (2026) · cites this paper as [104]
In the citing paper
“A minority of studies engaged clinicians to evaluate explanations [67,83,87,102,104], while most relied entirely on automated metrics or proxies.”
From Explainability of decoder-only clinical large language models: A scoping review (2026) · cites this paper as [104]
In the citing paper
“Only one [104] conducted a controlled user study comparing explanation formats.”
From Explainability of decoder-only clinical large language models: A scoping review (2026) · cites this paper as [104]
Citations
Cites
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
Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization
Selvaraju, R. R., Cogswell, M., Das, A., Vedantam, R., Parikh, D., & Batra, D. (2017) · 2017 IEEE International Conference on Computer Vision (ICCV) · doi:10.1109/iccv.2017.74 · Cited by 24,513
A Unified Approach to Interpreting Model Predictions
Lundberg, S., & Lee, S.-I. (2017) · arXiv · doi:10.48550/arxiv.1705.07874 · Cited by 23,707
Language Models are Few-Shot Learners
Brown, T. B., Mann, B., Ryder, N., Subbiah, M., Kaplan, J., Dhariwal, P., Neelakantan, A., Shyam, P., Sastry, G., Askell, A., et al. (2020) · arXiv · doi:10.48550/arxiv.2005.14165 · Cited by 23,372
Sequence to Sequence Learning with Neural Networks
Sutskever, I., Vinyals, O., & Le, Q. V. (2014) · arXiv · doi:10.48550/arxiv.1409.3215 · Cited by 17,766
"Why Should I Trust You?": Explaining the Predictions of Any Classifier
Ribeiro, M. T., Singh, S., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939778 · Cited by 17,555
Self-citationSentence-BERT: Sentence Embeddings using Siamese BERT-Networks
Reimers, N., & Gurevych, I. (2019) · Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP) · doi:10.18653/v1/d19-1410 · Cited by 12,778
Decoupled Weight Decay Regularization
Loshchilov, I., & Hutter, F. (2017) · arXiv · doi:10.48550/arxiv.1711.05101 · Cited by 12,522
Exploring the Limits of Transfer Learning with a Unified Text-to-Text Transformer
Raffel, C., Shazeer, N., Roberts, A., Lee, K., Narang, S., Matena, M., Zhou, Y., Li, W., & Liu, P. J. (2019) · arXiv · doi:10.48550/arxiv.1910.10683 · Cited by 10,929
From local explanations to global understanding with explainable AI for trees
Lundberg, S. M., Erion, G., Chen, H., DeGrave, A., Prutkin, J. M., Nair, B., Katz, R., Himmelfarb, J., Bansal, N., & Lee, S.-I. (2020) · Nature Machine Intelligence · doi:10.1038/s42256-019-0138-9 · Cited by 10,551
Research works printed but not linked: 5
- Slack, D., Rauschmayr, N. & Kenthapadi, K. Defuse: training more robust models through creation and correction of novel model errors. In NeurIPS 2021 Workshop on Explainable AI Approaches for Debugging and Diagnosis (2021).
- Lakkaraju, H., Slack, D., Chen, Y., Tan, C. & Sing, S. Rethinkingexplainability as a dialogue: a practitioner’s perspective. HAI Workshop @ NeurIPS (2022).
- Feldhus, N., Ravichandran, A. M. & Möller, S. Mediators: conversational agents explaining NLP model behavior. IJCAI-ECAI Workshop on Explainable Artificial Intelligence (XAI) (2022).
- Xie, S. M., Raghunathan, A., Liang, P. & Ma, T. An explanation of in-context learning as implicit bayesian inference. In International Conference on Learning Representations (2022).
- Guo, D., Tang, D., Duan, N., Zhou, M. & Yin, J. Dialog-to-action: conversational question answering over a large-scale knowledge base. In Proc. 32nd International Conference on Neural Information Processing Systems (eds Bengio, S. et al.) 2946–2955 (Curran Associates Inc., 2018).
Cited by
Sorted by citations
Feuerriegel, S., Hartmann, J., Janiesch, C., & Zschech, P. (2024) · Business & Information Systems Engineering · doi:10.1007/s12599-023-00834-7 · Cited by 2,267
“…could be, for example, to use generative AI for translating post hoc explanations derived from approaches like SHAP or LIME into more intuitive textual descriptions or generate user-friendly descriptions of models that are intrinsically inter-pretable (Slack et al. 2023; Zilker et al. 2023).”
Mersha, M. A., Lâm, K. N., Wood, J., AlShami, A. K., & Kalita, J. K. (2024) · Neurocomputing · doi:10.1016/j.neucom.2024.128111 · Cited by 346
Adversarial attacks and defenses in explainable artificial intelligence: A survey
Baniecki, H., & Biecek, P. (2024) · Information Fusion · doi:10.1016/j.inffus.2024.102303 · Cited by 234
Explainable Generative AI (GenXAI): a survey, conceptualization, and research agenda
Schneider, J. (2024) · Artificial Intelligence Review · doi:10.1007/s10462-024-10916-x · Cited by 214
Digital twins as global learning health and disease models for preventive and personalized medicine
Li, X., Loscalzo, J., Mahmud, A. K. M. F., Aly, D. M., Rzhetsky, A., Žitnik, M., & Benson, M. (2025) · Genome Medicine · doi:10.1186/s13073-025-01435-7 · Cited by 135
Metadata
- Authors
- Slack, D., Krishna, S., Lakkaraju, H., & Singh, S.
- Affiliations
- University of California, Irvine (US)
- Harvard University (US)
- Corresponding author
- Slack, Dylan
- Venue
- Nature Machine Intelligence
- Published
- 2023-07-27
- Volume
- 5
- Issue
- 8
- Pages
- 873-883
- ISSN
- 2522-5839 (ISSN-L)
- OpenAlex
- W4385302687
- Cited by
- 125
- References
- 80
- Citations by year
- 2026: 31 · 2025: 53 · 2024: 33 · 2023: 8
- FWCI
- 11.78
- Type
- journal article
- Language
- en
- Open access
- hybrid
- doi.org
- Locations
- Nature Machine Intelligence (publishedVersion, cc-by)
- Research Square (acceptedVersion, cc-by)
- Funders
- NSF | Directorate for Computer & Information Science & Engineering | Division of Information and Intelligent Systems (IIS-2040989, IIS-2046873, IIS-2008956, IIS-2008461)
- I was supported by a fellowship from the hasso plattner institute during the bulk of completing this work.
- Google, JP Morgan, Amazon, Harvard Data Science Initiative, D^3 institute at Harvard
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Explainable Artificial Intelligence (XAI)
- Physical Sciences › Computer Science › Software › Software Testing and Debugging Techniques
- Physical Sciences › Computer Science › Human-Computer Interaction › Usability and User Interface Design
- Keywords
- machine learning explainability, natural language dialogue systems, interactive explanations, conversational interfaces, model understanding, disease prediction model, healthcare workers
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
