EMNLP Findings2023
Learning Semantic Role Labeling from Compatible Label Sequences
Tao Li, Ghazaleh Kazeminejad, Susan Windisch Brown, Vivek Srikumar, Martha Palmer
Abstract
Semantic role labeling (SRL) has multiple disjoint label sets, e.g., VerbNet and PropBank. Creating these datasets is challenging, therefore a natural question is how to use each one to help the other. Prior work has shown that cross-task interaction helps, but only explored multitask learning so far. A common issue with multitask setup is that argument sequences are still separately decoded, running the risk of generating structurally inconsistent label sequences (as per lexicons like SEMLINK). In this paper, we eliminate such issue with a framework that jointly models VerbNet and PropBank labels as one sequence. In this setup, we show that enforcing SEMLINK constraints during decoding constantly improves the overall F1. With special input constructions, our joint model infers VerbNet arguments from given PropBank arguments with over 99 F1. For learning, we propose a constrained marginal model that learns with knowledge defined in SEMLINK to further benefit from the large amounts of PropBankonly data. On the joint benchmark based on CoNLL05, our models achieve state-of-the-art F1's, outperforming the prior best in-domain model by 3.5 (VerbNet) and 0.8 (PropBank). For out-of-domain generalization, our models surpass the prior best by 3.4 (VerbNet) and 0.2 (PropBank).
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Lineage
Builds on4
- Chain-of-Thought Prompting Elicits Reasoning in Large Language ModelsJason Wei, Xuezhi Wang, Dale Schuurmans, Maarten Bosma et al.NeurIPS 2022 · 22,562 citations
- Compositional Semantic Parsing with Large Language ModelsAndrew Drozdov, Nathanael Schärli, Ekin Akyürek, Nathan Scales et al.ICLR 2023 · 39 citations
- Structured Tuning for Semantic Role LabelingTao Li, Parth Anand Jawale, Martha Palmer, Vivek SrikumarACL 2020 · 2 citations
- A matter of framing: The impact of linguistic formalism on probing resultsIlia Kuznetsov, Iryna GurevychEMNLP 2020 · 1 citation
Related papers
- Exploring Non-Verbal Predicates in Semantic Role Labeling: Challenges and OpportunitiesRiccardo Orlando, Simone Conia, Roberto NavigliACL Findings 2023 · 3 citations
- UniteD-SRL: A Unified Dataset for Span- and Dependency-Based Multilingual and Cross-Lingual Semantic Role LabelingRocco Tripodi, Simone Conia, Roberto NavigliEMNLP Findings 2021 · 2 citations
- Semantic Role Labeling as Syntactic Dependency ParsingTianze Shi, Igor Malioutov, Ozan IrsoyEMNLP 2020 · 15 citations
- On the Benefit of Syntactic Supervision for Cross-lingual Transfer in Semantic Role LabelingZhisong Zhang, Emma Strubell, Eduard H. HovyEMNLP 2021 · 1 citation
- End-to-end Semantic Role Labeling with Neural Transition-based ModelHao Fei, Meishan Zhang, Bobo Li, Donghong JiAAAI 2021 · 42 citations
