P-Adapters: Robustly Extracting Factual Information from Language Models with Diverse Prompts

TL;DR

This paper introduces P-Adapters, lightweight models that improve the consistency and accuracy of factual information extraction from LLMs across diverse prompts without requiring additional annotations.

Contribution

The work proposes P-Adapters, a novel lightweight approach that enhances factual extraction consistency from LLMs, outperforming complex MoE models without extra annotation needs.

Findings

P-Adapters achieve 12-26% improvement in precision.

P-Adapters show 36-50% improvement in consistency.

Access to original prompt embeddings is key to success.

Abstract

Recent work (e.g. LAMA (Petroni et al., 2019)) has found that the quality of the factual information extracted from Large Language Models (LLMs) depends on the prompts used to query them. This inconsistency is problematic because different users will query LLMs for the same information using different wording, but should receive the same, accurate responses regardless. In this work we aim to address this shortcoming by introducing P-Adapters: lightweight models that sit between the embedding layer and first attention layer of LLMs. They take LLM embeddings as input and output continuous prompts that are used to query the LLM. Additionally, we investigate Mixture of Experts (MoE) models that learn a set of continuous prompts ("experts") and select one to query the LLM. They require a separate classifier trained on human-annotated data to map natural language prompts to the continuous ones. P-Adapters perform comparably to the more complex MoE models in extracting factual information from BERT and RoBERTa while eliminating the need for additional annotations. P-Adapters show between 12-26% absolute improvement in precision and 36-50% absolute improvement in consistency over a baseline of only using natural language queries. Finally, we investigate what makes P-Adapters successful and conclude that a significant factor is access to the LLM's embeddings of the original natural language prompt, particularly the subject of the entity pair being queried.