InternalInspector $I^2$: Robust Confidence Estimation in LLMs through Internal States

TL;DR

InternalInspector is a new framework that improves confidence estimation in LLMs by analyzing all internal states across layers, leading to better detection of hallucinations and inaccuracies.

Contribution

It introduces a comprehensive internal state analysis using contrastive learning, surpassing existing methods in confidence calibration and hallucination detection.

Findings

Higher accuracy in confidence estimation across tasks

Lower calibration error compared to existing methods

Superior hallucination detection performance

Abstract

Despite their vast capabilities, Large Language Models (LLMs) often struggle with generating reliable outputs, frequently producing high-confidence inaccuracies known as hallucinations. Addressing this challenge, our research introduces InternalInspector, a novel framework designed to enhance confidence estimation in LLMs by leveraging contrastive learning on internal states including attention states, feed-forward states, and activation states of all layers. Unlike existing methods that primarily focus on the final activation state, InternalInspector conducts a comprehensive analysis across all internal states of every layer to accurately identify both correct and incorrect prediction processes. By benchmarking InternalInspector against existing confidence estimation methods across various natural language understanding and generation tasks, including factual question answering, commonsense reasoning, and reading comprehension, InternalInspector achieves significantly higher accuracy in aligning the estimated confidence scores with the correctness of the LLM's predictions and lower calibration error. Furthermore, InternalInspector excels at HaluEval, a hallucination detection benchmark, outperforming other internal-based confidence estimation methods in this task.