PCS: Perceived Confidence Scoring of Black Box LLMs with Metamorphic Relations
This addresses the need for better confidence estimation in black-box LLMs for data annotation tasks, offering an incremental improvement over existing methods.
The paper tackles the problem of suboptimal performance in zero-shot LLMs for textual classification tasks by introducing a perceived confidence scoring technique based on metamorphic relations, which improves performance by 9.3% for single LLMs and 5.8% in majority-voting setups.
Zero-shot LLMs are now also used for textual classification tasks, e.g., sentiment and bias detection in a sentence or article. However, their performance can be suboptimal in such data annotation tasks. We introduce a novel technique that evaluates an LLM's confidence for classifying a textual input by leveraging Metamorphic Relations (MRs). The MRs generate semantically equivalent yet textually divergent versions of the input. Following the principles of Metamorphic Testing (MT), the mutated versions are expected to have annotation labels similar to the input. By analyzing the consistency of an LLM's responses across these variations, we compute a perceived confidence score (PCS) based on the frequency of the predicted labels. PCS can be used for both single and multiple LLM settings (e.g., when multiple LLMs are vetted in a majority-voting setup). Empirical evaluation shows that our PCS-based approach improves the performance of zero-shot LLMs by 9.3% in textual classification tasks. When multiple LLMs are used in a majority-voting setup, we obtain a performance boost of 5.8% with PCS.