# Towards Improving Selective Prediction Ability of NLP Systems

arXiv: https://arxiv.org/abs/2008.09371 · HTML: https://mlsanity.com/paper/2008.09371/

## Key facts

- Cited by 20 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 21 Aug 2020.

## Built on by (most cited first, 20 of 20)

- [Generating with Confidence: Uncertainty Quantification for Black-box Large Language Models](https://mlsanity.com/paper/2305.19187/) (2023)
- [Uncertainty Quantification with Pre-trained Language Models: A Large-Scale Empirical Analysis](https://mlsanity.com/paper/2210.04714/) (2022)
- [The Art of Defending: A Systematic Evaluation and Analysis of LLM Defense Strategies on Safety and Over-Defensiveness](https://mlsanity.com/paper/2401.00287/) (2023)
- [Uncertainty in Natural Language Processing: Sources, Quantification, and Applications](https://mlsanity.com/paper/2306.04459/) (2023)
- [Overcoming Common Flaws in the Evaluation of Selective Classification Systems](https://mlsanity.com/paper/2407.01032/) (2024)
- [Learning to Reject with a Fixed Predictor: Application to Decontextualization](https://mlsanity.com/paper/2301.09044/) (2023)
- [Post-Abstention: Towards Reliably Re-Attempting the Abstained Instances in QA](https://mlsanity.com/paper/2305.01812/) (2023)
- [Contextualized Sequence Likelihood: Enhanced Confidence Scores for Natural Language Generation](https://mlsanity.com/paper/2406.01806/) (2024)
- [Measuring the Robustness of NLP Models to Domain Shifts](https://mlsanity.com/paper/2306.00168/) (2023)
- [Can Open-Domain QA Reader Utilize External Knowledge Efficiently like Humans?](https://mlsanity.com/paper/2211.12707/) (2022)
- [Let the Model Decide its Curriculum for Multitask Learning](https://mlsanity.com/paper/2205.09898/) (2022)
- [Ambiguity Meets Uncertainty: Investigating Uncertainty Estimation for Word Sense Disambiguation](https://mlsanity.com/paper/2305.13119/) (2023)
- [Discover, Explanation, Improvement: An Automatic Slice Detection Framework for Natural Language Processing](https://mlsanity.com/paper/2211.04476/) (2022)
- [Can NLP Models 'Identify', 'Distinguish', and 'Justify' Questions that Don't have a Definitive Answer?](https://mlsanity.com/paper/2309.04635/) (2023)
- [A Unified Evaluation Framework for Novelty Detection and Accommodation in NLP with an Instantiation in Authorship Attribution](https://mlsanity.com/paper/2305.05079/) (2023)
- [Out-of-Distribution Generalization in Text Classification: Past, Present, and Future](https://mlsanity.com/paper/2305.14104/) (2023)
- [The Craft of Selective Prediction: Towards Reliable Case Outcome Classification -- An Empirical Study on European Court of Human Rights Cases](https://mlsanity.com/paper/2409.18645/) (2024)
- [ECUASn: A family of metrics for principled evaluation of uncertainty-augmented systems](https://mlsanity.com/paper/2605.20490/) (2026)
- [Know2Guess: A Contamination-Aware Multi-Zone Benchmark for Knowledge-Boundary Evaluation in Large Language Models](https://mlsanity.com/paper/2606.26101/) (2026)
- [Strong yet Controllable LLM Abstention without Learning to Abstain](https://mlsanity.com/paper/2610.16318/) (2026)

## Abstract (arXiv)

It's better to say "I can't answer" than to answer incorrectly. This selective prediction ability is crucial for NLP systems to be reliably deployed in real-world applications. Prior work has shown that existing selective prediction techniques fail to perform well, especially in the out-of-domain setting. In this work, we propose a method that improves probability estimates of models by calibrating them using prediction confidence and difficulty score of instances. Using these two signals, we first annotate held-out instances and then train a calibrator to predict the likelihood of correctness of the model's prediction. We instantiate our method with Natural Language Inference (NLI) and Duplicate Detection (DD) tasks and evaluate it in both In-Domain (IID) and Out-of-Domain (OOD) settings. In (IID, OOD) settings, we show that the representations learned by our calibrator result in an improvement of (15.81%, 5.64%) and (6.19%, 13.9%) over 'MaxProb' -- a selective prediction baseline -- on NLI and DD tasks respectively.