Investigating Selective Prediction Approaches Across Several Tasks in IID, OOD, and Adversarial Settings

By Neeraj Varshney, Swaroop Mishra, Chitta Baral. 2022

68 citations recorded by the source corpus.

In order to equip NLP systems with ‘selective prediction’ capability, several task-specific approaches have been proposed. However, which approaches work best across tasks or even if they consistently outperform the simplest baseline MaxProb remains to be explored. To this end, we systematically study selective prediction in a large-scale setup of 17 datasets across several NLP tasks. Through comprehensive experiments under in-domain (IID), out-of-domain (OOD), and adversarial (ADV) settings, we show that despite leveraging additional resources (held-out data/computation), none of the existing approaches consistently and considerably outperforms MaxProb in all three settings. Furthermore, their performance does not translate well across tasks. For instance, Monte-Carlo Dropout outperforms all other approaches on Duplicate Detection datasets but does not fare well on NLI datasets, especially in the OOD setting. Thus, we recommend that future selective prediction approaches should be evaluated across tasks and settings for reliable estimation of their capabilities.

Subfields: Safety, Robustness & Interpretability

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety, Robustness & Interpretability

1. Shubhendu Trivedi
2. Jimeng Sun
3. Zhen Lin

Citations in the corpus

• Generating with Confidence: Uncertainty Quantification for Black-box Large Language Models (2023, background): “Recent works have explored selective classification in NLP tasks (Varshney et al., 2022a;b).”