The DOI matches its Crossref record; 72 of the 72 references it cites were found in a registry or an index.
How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering
Open accessAbstract
Recent works have shown that language models (LM) capture different types of knowledge regarding facts or common sense. However, because no model is perfect, they still fail to provide appropriate answers in many cases. In this paper, we ask the question, “How can we know when language models know, with confidence, the answer to a particular query?” We examine this question from the point of view of calibration, the property of a probabilistic model’s predicted probabilities actually being well correlated with the probabilities of correctness. We examine three strong generative models—T5, BART, and GPT-2—and study whether their probabilities on QA tasks are well calibrated, finding the answer is a relatively emphatic no. We then examine methods to calibrate such models to make their confidence scores correlate better with the likelihood of correctness through fine-tuning, post-hoc probability modification, or adjustment of the predicted outputs or inputs. Experiments on a diverse range of datasets demonstrate the effectiveness of our methods. We also perform analysis to study the strengths and limitations of these methods, shedding light on further improvements that may be made in methods for calibrating LMs. We have released the code at https://github.com/jzbjyb/lm-calibration.
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?
The cited source doesn’t support the claim as written.
In the citing paper
“Another line of work that sheds light on the nature of hallucinations shows that LMs know when they’re hallucinating [1,3,18,20].”
From Calibrated Language Models Must Hallucinate (2024) · cites this paper as [18]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe cited source doesn’t support the claim as written.
In the citing paper
“Jiang et al. [51] observed the probabilities of pre-trained LMs (e.g., BART, T5, and GPT-2) on QA tasks are well calibrated.”
From Pre-train, Prompt, and Predict: A Systematic Survey of Prompting Methods in Natural Language Processing (2023) · cites this paper as [51]
Checked by a reasoning model — may be wrong.
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“Past work on uncertainty for LLMs has focused on simpler settings, such as classifiers16,17 and regressors18,19, whereas the most exciting applications of LLMs relate to free-form generations.”
From Detecting hallucinations in large language models using semantic entropy (2024) · cites this paper as [16]
Experimental — accuracy not yet measuredThe source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“find only a weak correlation between the log-likelihood models assign to their answer and the answer's correctness.”
From Semantic Uncertainty: Linguistic Invariances for Uncertainty Estimation in Natural Language Generation (2023) · cites this paper as Jiang et al. (2021)
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“More complex decision-making is currently bottlenecked by issues such as over-confidence and miscalibration (Jiang et al., 2021; Braverman et al., 2020; Chen et al., 2022), misalignment with human values or bias (Liang et al., 2021; Feng et al., 2023), hallucinations in self-evaluation (Shinn et al., 2023), and lack of human-in-the-loop mechanisms in face of uncertainties (Nguyen et al., 2022a; R”
From Cognitive Architectures for Language Agents (2023) · cites this paper as Jiang et al., 2021
8 more claims about this paper
Create a free account to see every claim other papers make about this one.
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
Deep Residual Learning for Image Recognition
He, K., Zhang, X., Ren, S., & Sun, J. (2016) · 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) · doi:10.1109/cvpr.2016.90 · Cited by 235,474 · cited 2×
“We found that larger LMs usually achieve both higher accuracy and better calibration performance, which is contradictory to the observation in image clas-sification (Guo et al., 2017) where larger models such as ResNet (He et al., 2016) are no longer well calibrated compared to smaller models like LeNet (Lecun et al.,…”
MethodVaswani, A., Shazeer, N., Parmar, N., Uszkoreit, J., Jones, L., Gomez, A. N., Kaiser, L., & Polosukhin, I. (2017) · arXiv · doi:10.48550/arxiv.1706.03762 · Cited by 120,404 · cited 1×
“The T5 model is a sequence-to-sequence model with both encoder and decoder using trans-formers (Vaswani et al., 2017), and the largest version has 11 billion parameters, allowing it to realize state-of-the-art performance on tasks such as question answering and natural language understanding…”
BackgroundGradient-based learning applied to document recognition
Lecun, Y., Bottou, L., Bengio, Y., & Haffner, P. (1998) · Proceedings of the IEEE · doi:10.1109/5.726791 · Cited by 58,805 · cited 2×
“We found that larger LMs usually achieve both higher accuracy and better calibration performance, which is contradictory to the observation in image clas-sification (Guo et al., 2017) where larger models such as ResNet (He et al., 2016) are no longer well calibrated compared to smaller models like LeNet (Lecun et al.,…”
MethodXGBoost: A Scalable Tree Boosting System
Chen, T., & Guestrin, C. (2016) · Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining · doi:10.1145/2939672.2939785 · Cited by 54,569 · cited 2×
“We follow previous works (Jagannatha and Yu, 2020; Dong et al., 2018) and use gradient boosted decision trees (Chen and Guestrin, 2016) as our regres-sor to estimate the confidence based on features.”
MethodInfluentialBERT: Pre-training of Deep Bidirectional Transformers for Language Understanding
Devlin, J., Chang, M.-W., Lee, K., & Toutanova, K. (2019) · Proceedings of the 2019 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, Volume 1 (Long and Short Papers) · doi:10.18653/v1/n19-1423 · Cited by 54,325 · cited 3×
“LMs are now a ubiquitous tool in not only natu-ral language generation, but also natural language understanding (NLU), where they are largely used for unsupervised representation learning in pre-trained models such as BERT (Devlin et al., 2019).”
MethodInfluentialLanguage 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
Kaugeranna, R., Kaugeranna, E., & Aion, (. S. 4. 6. (2023) · DROPS (Schloss Dagstuhl – Leibniz Center for Informatics) · doi:10.4230/lipics.giscience.2023.43 · Cited by 13,690
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 · cited 3×
“No-tably, large models trained on massive crawls of internet text (such as T5 (Raffel et al., 2019) and GPT-3 (Brown et al., 2020)) have been shown to be able to perform quite sophisticated knowledge-based tasks simply through prompting the model to predict the next words given a particular cue.”
BackgroundMethodScaling Laws for Neural Language Models
Kaplan, J., McCandlish, S., Henighan, T., Brown, T. B., Chess, B., Child, R., Gray, S., Radford, A., Wu, J., & Amodei, D. (2020) · arXiv · doi:10.48550/arxiv.2001.08361 · Cited by 9,550 · cited 1×
“Unlike ResNet trained on CIFAR-100, the training of LMs is not bottlenecked by the dataset, and larger LMs have a stronger capacity to model text distribution and memorize facts, which leads to better calibration performance overall (Kaplan et al., 2020).”
MethodLewis, M., Liu, Y., Goyal, N., Ghazvininejad, M., Mohamed, A., Levy, O., Stoyanov, V., & Zettlemoyer, L. (2020) · Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics · doi:10.18653/v1/2020.acl-main.703 · Cited by 8,821
Research works printed but not linked: 4
- Hewitt, 2019, A structural probe for finding syntax in word representations, 4129
- Ng, 2019, Facebook fair’s WMT19 news translation task submission, 314
- Quach, 2020, Researchers made an OpenAI GPT-3 medical chatbot as an experiment. It told a mock patient to kill themselves, The Register
- Radford, 2019, Language models are unsupervised multitask learners, OpenAI Blog, 1
Cited by
Liu, P., Yuan, W., Fu, J., Jiang, Z., Hayashi, H., & Neubig, G. (2023) · ACM Computing Surveys · doi:10.1145/3560815 · Cited by 4,361
“• Answer Paraphrasing Thesemethods start with an initial answer spaceZ, and then use paraphrasing to expand this answer space to broaden its coverage [51].”
BackgroundMethodInfluentialSelf-citationRecent Advances in Natural Language Processing via Large Pre-trained Language Models: A Survey
Min, B., Ross, H., Sulem, E., Veyseh, A. P. B., Nguyen, T. H., Sainz, O., Agirre, E., Heintz, I., & Roth, D. (2024) · ACM Computing Surveys · doi:10.1145/3605943 · Cited by 1,387
“The prompts can be combined using a uniform average [75, 166, 210] or a weighted average [75, 143, 166, 167].”
BackgroundMethodInfluentialDetecting hallucinations in large language models using semantic entropy
Farquhar, S., Kossen, J., Kuhn, L., & Gal, Y. (2024) · Nature · doi:10.1038/s41586-024-07421-0 · Cited by 934
Siren's Song in the AI Ocean: A Survey on Hallucination in Large Language Models
Zhang, Y., Li, Y., Cui, L., Cai, D., Liu, L., Fu, T., Huang, X., Zhao, E., Zhang, Y., Chen, Y., et al. (2025) · Computational Linguistics · doi:10.1162/coli.a.16 · Cited by 534
Calibrate Before Use: Improving Few-Shot Performance of Language Models
Zhao, T. Z., Wallace, E., Feng, S., Klein, D., & Singh, S. (2021) · arXiv · doi:10.48550/arxiv.2102.09690 · Cited by 513
Metadata
- Authors
- Jiang, Z., Araki, J., Ding, H., & Neubig, G.
- Affiliations
- Carnegie Mellon University (US)
- Robert Bosch (United States) (US)
- Corresponding author
- Jiang, Zhengbao
- Araki, Jun
- Ding, Haibo
- Neubig, Graham
- Venue
- Transactions of the Association for Computational Linguistics
- Publisher
- MIT Press
- Published
- 2021
- Volume
- 9
- Pages
- 962-977
- ISSN
- 2307-387X (ISSN-L)
- OpenAlex
- W3199958362
- Cited by
- 177
- References
- 71
- Citations by year
- 2026: 43 · 2025: 49 · 2024: 32 · 2023: 37 · 2022: 14 · 2021: 2
- FWCI
- 7.48
- Type
- journal article
- Language
- en
- Open access
- diamond
- doi.org
- Locations
- Transactions of the Association for Computational Linguistics (publishedVersion, cc-by)
- DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
- Topics
- Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- Physical Sciences › Computer Science › Artificial Intelligence › Adversarial Robustness in Machine Learning
- Physical Sciences › Computer Science › Artificial Intelligence › Natural Language Processing Techniques
- Keywords
- question answering, confidence calibration, probabilistic reasoning, T5, BART, GPT-2, fine-tuning, knowledge capture, factual knowledge, commonsense knowledge
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering. https://app.cochat.ai/papers/10.1162/tacl_a_00407
- The paper
- Jiang, Z., Araki, J., Ding, H., & Neubig, G. (2021). How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering. Transactions of the Association for Computational Linguistics. https://doi.org/10.1162/tacl_a_00407
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.