Schema-Constrained Document-Level Event Argument Extraction with Lightweight LLM Fine-Tuning
Computation and Language
Summary
The authors studied how well medium-sized open large language models (LLMs) can extract event information from entire documents while following strict rules. They enhanced the models by adding role details in prompts, fine-tuning them efficiently, and using smart decoding methods to make outputs more accurate and consistent. Their best fine-tuned model performed better than previous GPT models on a benchmark dataset called MAVEN-ARG. They also shared their code so others can try their approach.
Authors
Roberto Pietrantuono, Antonio Guerriero, Pouya Sattari
Abstract
Event Argument Extraction (EAE) converts documents into structured event records by identifying argument spans and assigning them schema-defined roles. Document-level EAE is challenging due to long-range dependencies between triggers and arguments, cross-sentence context, and strict role constraints, which often lead to boundary errors, uncertainty in roles, and inconsistencies with restricted schemas. In this paper, we study whether mid-sized open LLMs can perform schema-constrained EAE reliably at the document level on MAVEN-ARG. Our approach combines (i) role-set injection in prompts for schema compliance, (ii) parameter-efficient supervised fine-tuning (LoRA) using the same JSON-only interface used at inference, and (iii) deterministic decoding with post-processing that validates JSON, filters invalid roles, de-duplicates arguments, and aligns spans to the document window. Under the official MAVEN-ARG evaluator, fine-tuned mid-sized open models outperform previously reported GPT baselines across mention, entity-coreference, and event-coreference evaluations; our best model (Phi-4, 14B) reaches 42.39\% F1 at the event-coreference level. Code to reproduce experiments is publicly available at https://github.com/dessertlab/EAE/.