Teaching Models to Express Their Uncertainty in Words

• URL: http://arxiv.org/abs/2205.14334v1
• Date: Sat, 28 May 2022 05:02:31 GMT
• Title: Teaching Models to Express Their Uncertainty in Words
• Authors: Stephanie Lin, Jacob Hilton, Owain Evans
• Abstract summary: We show that a GPT-3 model can learn to express uncertainty about its own answers in natural language. This is the first time a model has been shown to express calibrated uncertainty about its own answers in natural language.
• Score: 6.356472059420951
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: We show that a GPT-3 model can learn to express uncertainty about its own answers in natural language -- without use of model logits. When given a question, the model generates both an answer and a level of confidence (e.g. "90% confidence" or "high confidence"). These levels map to probabilities that are well calibrated. The model also remains moderately calibrated under distribution shift, and is sensitive to uncertainty in its own answers, rather than imitating human examples. To our knowledge, this is the first time a model has been shown to express calibrated uncertainty about its own answers in natural language. For testing calibration, we introduce the CalibratedMath suite of tasks. We compare the calibration of uncertainty expressed in words ("verbalized probability") to uncertainty extracted from model logits. Both kinds of uncertainty are capable of generalizing calibration under distribution shift. We also provide evidence that GPT-3's ability to generalize calibration depends on pre-trained latent representations that correlate with epistemic uncertainty over its answers.

Related papers

• LACIE: Listener-Aware Finetuning for Confidence Calibration in Large Language Models [69.68379406317682]
  We introduce a listener-aware finetuning method (LACIE) to calibrate implicit and explicit confidence markers. We show that LACIE models the listener, considering not only whether an answer is right, but whether it will be accepted by a listener. We find that training with LACIE results in 47% fewer incorrect answers being accepted while maintaining the same level of acceptance for correct answers.
  arXiv  Detail & Related papers  (2024-05-31T17:16:38Z)
• Selective Learning: Towards Robust Calibration with Dynamic Regularization [79.92633587914659]
  Miscalibration in deep learning refers to there is a discrepancy between the predicted confidence and performance. We introduce Dynamic Regularization (DReg) which aims to learn what should be learned during training thereby circumventing the confidence adjusting trade-off.
  arXiv  Detail & Related papers  (2024-02-13T11:25:20Z)
• Calibrated Uncertainty Quantification for Operator Learning via Conformal Prediction [95.75771195913046]
  We propose a risk-controlling quantile neural operator, a distribution-free, finite-sample functional calibration conformal prediction method. We provide a theoretical calibration guarantee on the coverage rate, defined as the expected percentage of points on the function domain. Empirical results on a 2D Darcy flow and a 3D car surface pressure prediction task validate our theoretical results.
  arXiv  Detail & Related papers  (2024-02-02T23:43:28Z)
• Improving the Reliability of Large Language Models by Leveraging Uncertainty-Aware In-Context Learning [76.98542249776257]
  Large-scale language models often face the challenge of "hallucination" We introduce an uncertainty-aware in-context learning framework to empower the model to enhance or reject its output in response to uncertainty.
  arXiv  Detail & Related papers  (2023-10-07T12:06:53Z)
• Trusting Language Models in Education [1.2578554943276923]
  We propose to use an XGBoost on top of BERT to output the corrected probabilities. Our hypothesis is that the level of uncertainty contained in the flow of attention is related to the quality of the model's response itself.
  arXiv  Detail & Related papers  (2023-08-07T18:27:54Z)
• Calibration Meets Explanation: A Simple and Effective Approach for Model Confidence Estimates [21.017890579840145]
  We propose a method named CME that leverages model explanations to make the model less confident with non-inductive attributions. We conduct extensive experiments on six datasets with two popular pre-trained language models. Our findings highlight that model explanations can help calibrate posterior estimates.
  arXiv  Detail & Related papers  (2022-11-06T06:17:21Z)
• Why Calibration Error is Wrong Given Model Uncertainty: Using Posterior Predictive Checks with Deep Learning [0.0]
  We show how calibration error and its variants are almost always incorrect to use given model uncertainty. We show how this mistake can lead to trust in bad models and mistrust in good models.
  arXiv  Detail & Related papers  (2021-12-02T18:26:30Z)
• How Can We Know When Language Models Know? On the Calibration of Language Models for Question Answering [80.82194311274694]
  We examine the question "how can we know when language models know, with confidence, the answer to a particular query?" We examine three strong generative models -- T5, BART, and GPT-2 -- and study whether their probabilities on QA tasks are well calibrated. We then examine methods to calibrate such models to make their confidence scores correlate better with the likelihood of correctness.
  arXiv  Detail & Related papers  (2020-12-02T03:53:13Z)
• Calibration of Pre-trained Transformers [55.57083429195445]
  We focus on BERT and RoBERTa in this work, and analyze their calibration across three tasks: natural language inference, paraphrase detection, and commonsense reasoning. We show that: (1) when used out-of-the-box, pre-trained models are calibrated in-domain, and compared to baselines, their calibration error out-of-domain can be as much as 3.5x lower; (2) temperature scaling is effective at further reducing calibration error in-domain, and using label smoothing to deliberately increase empirical uncertainty helps calibrate posteriors out-of-domain.
  arXiv  Detail & Related papers  (2020-03-17T18:58:44Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.