# Encoding Syntactic Constituency Paths for Frame-Semantic Parsing with Graph Convolutional Networks

arXiv: https://arxiv.org/abs/2011.13210 · HTML: https://mlsanity.com/paper/2011.13210/

## Key facts

- Cited by 7 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 26 Nov 2020.

## Built on by (most cited first, 7 of 7)

- [A Double-Graph Based Framework for Frame Semantic Parsing](https://mlsanity.com/paper/2206.09158/) (2022)
- [A Graph-Based Neural Model for End-to-End Frame Semantic Parsing](https://mlsanity.com/paper/2109.12319/) (2021)
- [Can LLMs Extract Frame-Semantic Arguments?](https://mlsanity.com/paper/2502.12516/) (2025)
- [Query Your Model with Definitions in FrameNet: An Effective Method for Frame Semantic Role Labeling](https://mlsanity.com/paper/2212.02036/) (2022)
- [Exploring In-Context Learning for Frame-Semantic Parsing](https://mlsanity.com/paper/2507.23082/) (2025)
- [Knowledge-augmented Frame Semantic Parsing with Hybrid Prompt-tuning](https://mlsanity.com/paper/2303.14375/) (2023)
- [Lexicalized Constituency Parsing for Middle Dutch: Low-resource Training and Cross-Domain Generalization](https://mlsanity.com/paper/2601.07008/) (2026)

## Abstract (arXiv)

We study the problem of integrating syntactic information from constituency trees into a neural model in Frame-semantic parsing sub-tasks, namely Target Identification (TI), FrameIdentification (FI), and Semantic Role Labeling (SRL). We use a Graph Convolutional Network to learn specific representations of constituents, such that each constituent is profiled as the production grammar rule it corresponds to. We leverage these representations to build syntactic features for each word in a sentence, computed as the sum of all the constituents on the path between a word and a task-specific node in the tree, e.g. the target predicate for SRL. Our approach improves state-of-the-art results on the TI and SRL of ~1%and~3.5% points, respectively (+2.5% additional points are gained with BERT as input), when tested on FrameNet 1.5, while yielding comparable results on the CoNLL05 dataset to other syntax-aware systems.