Continuous Risk Prediction

TL;DR

Diana is a novel architecture-based lifelong QA framework that uses hierarchical prompts to improve knowledge retention, generalization, and handling of unseen tasks in question-answering models.

Contribution

The paper introduces Diana, a dynamic prompt-enhanced architecture for lifelong QA, incorporating multiple prompt types for better knowledge modeling and transfer, especially for unseen tasks.

Findings

Achieves state-of-the-art lifelong QA performance.

Significantly improves handling of unseen tasks.

Enhances generalization through hierarchical prompts.

Abstract

Lifelong learning (LL) capabilities are essential for QA models to excel in real-world applications, and architecture-based LL approaches have proven to be a promising direction for achieving this goal. However, adapting existing methods to QA tasks is far from straightforward. Many prior approaches either rely on access to task identities during testing or fail to adequately model samples from unseen tasks, which limits their practical applicability. To overcome these limitations, we introduce Diana , a novel \underline{d}ynam\underline{i}c \underline{a}rchitecture-based lifelo\underline{n}g Q\underline{A} framework designed to learn a sequence of QA tasks using a prompt-enhanced language model.Diana leverages four hierarchically structured types of prompts to capture QA knowledge at multiple levels of granularity. Task-level prompts are specifically designed to encode task-specific knowledge, ensuring strong lifelong learning performance. Meanwhile, instance-level prompts are utilized to capture shared knowledge across diverse input samples, enhancing the model's generalization capabilities. Additionally, Diana incorporates dedicated prompts to explicitly handle unseen tasks and introduces a set of prompt key vectors that facilitate efficient knowledge transfer and sharing between tasks. Through extensive experimentation, we demonstrate that Diana achieves state-of-the-art performance among lifelong QA models, with particularly notable improvements in its ability to handle previously unseen tasks. This makes Diana a significant advancement in the field of lifelong learning for question-answering systems.