Adversarial Topic-aware Prompt-tuning for Cross-topic Automated Essay Scoring

TL;DR

This paper introduces ATOP, a novel adversarial prompt-tuning method that jointly learns topic-shared and topic-specific features to improve cross-topic automated essay scoring, outperforming existing methods on benchmark datasets.

Contribution

The paper proposes a new adversarial prompt-tuning approach that models both shared and specific topic features, enhancing transferability and scoring accuracy in cross-topic AES.

Findings

ATOP significantly outperforms state-of-the-art methods on the ASAP++ dataset.

Incorporating adversarial training improves robustness of topic-shared feature learning.

Using pseudo-labels guides the learning of topic-specific prompts effectively.

Abstract

Cross-topic automated essay scoring (AES) aims to develop a transferable model capable of effectively evaluating essays on a target topic. A significant challenge in this domain arises from the inherent discrepancies between topics. While existing methods predominantly focus on extracting topic-shared features through distribution alignment of source and target topics, they often neglect topic-specific features, limiting their ability to assess critical traits such as topic adherence. To address this limitation, we propose an Adversarial TOpic-aware Prompt-tuning (ATOP), a novel method that jointly learns topic-shared and topic-specific features to improve cross-topic AES. ATOP achieves this by optimizing a learnable topic-aware prompt--comprising both shared and specific components--to elicit relevant knowledge from pre-trained language models (PLMs). To enhance the robustness of topic-shared prompt learning and mitigate feature scale sensitivity introduced by topic alignment, we incorporate adversarial training within a unified regression and classification framework. In addition, we employ a neighbor-based classifier to model the local structure of essay representations and generate pseudo-labels for target-topic essays. These pseudo-labels are then used to guide the supervised learning of topic-specific prompts tailored to the target topic. Extensive experiments on the publicly available ASAP++ dataset demonstrate that ATOP significantly outperforms existing state-of-the-art methods in both holistic and multi-trait essay scoring. The implementation of our method is publicly available at: https://anonymous.4open.science/r/ATOP-A271.