DEGAP: Dual Event-Guided Adaptive Prefixes for Templated-Based Event Argument Extraction with Slot Querying

2024-05-22

International Conference on Computational Linguistics

AI Summary

To address template isolation and retrieval noise in event argument extraction (EAE), this paper proposes a dual-event-guided adaptive prefix mechanism. It introduces instance-aware and template-aware learnable prompt vectors to explicitly model latent inter-type relationships, eliminating reliance on external retrieval. A dual-prefix collaborative modeling framework with event-aware adaptive gating enables end-to-end learning of cross-template dependencies. Furthermore, template-driven slot querying is integrated with Prefix Tuning within the Transformer architecture to achieve joint argument extraction. Evaluated on four standard benchmarks—ACE05, RAMS, WikiEvents, and MLEE—the method achieves state-of-the-art performance. It notably enhances cross-event generalization and demonstrates superior robustness under low-resource conditions.

Technology Category

• Natural Language Processing: Discourse, Pragmatics & Argument Mining
• Knowledge Representation and Reasoning: Argumentation
• Cognitive Modeling & Cognitive Systems: Conceptual Inference and Reasoning

Application Category

• Search and Retrieval-Augmented AI: Retrieval-Augmented Generation (RAG) and multi-modal RAG
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking

Abstract

Recent advancements in event argument extraction (EAE) involve incorporating useful auxiliary information into models during training and inference, such as retrieved instances and event templates. These methods face two challenges: (1) the retrieval results may be irrelevant and (2) templates are developed independently for each event without considering their possible relationship. In this work, we propose DEGAP to address these challenges through a simple yet effective components: dual prefixes, i.e. learnable prompt vectors, where the instance-oriented prefix and template-oriented prefix are trained to learn information from different event instances and templates. Additionally, we propose an event-guided adaptive gating mechanism, which can adaptively leverage possible connections between different events and thus capture relevant information from the prefix. Finally, these event-guided prefixes provide relevant information as cues to EAE model without retrieval. Extensive experiments demonstrate that our method achieves new state-of-the-art performance on four datasets (ACE05, RAMS, WIKIEVENTS, and MLEE). Further analysis shows the impact of different components.

Problem

• Addresses irrelevant retrieval results in event argument extraction
• Resolves independent template design ignoring event relationships
• Enhances EAE with adaptive dual prefixes and gating

Innovation

• Dual prefixes for instance and template learning
• Event-guided adaptive gating mechanism
• No retrieval needed for relevant information