# Improved Semantic Role Labeling using Parameterized Neighborhood Memory Adaptation

arXiv: https://arxiv.org/abs/2011.14459 · HTML: https://mlsanity.com/paper/2011.14459/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Nov 2020.

## Built on by (most cited first, 6 of 6)

- [NL-Augmenter: A Framework for Task-Sensitive Natural Language Augmentation](https://mlsanity.com/paper/2112.02721/) (2021)
- [Semantic Role Labeling as Dependency Parsing: Exploring Latent Tree Structures Inside Arguments](https://mlsanity.com/paper/2110.06865/) (2021)
- [Syntax-Aware Graph-to-Graph Transformer for Semantic Role Labelling](https://mlsanity.com/paper/2104.07704/) (2021)
- [An MRC Framework for Semantic Role Labeling](https://mlsanity.com/paper/2109.06660/) (2021)
- [Fast and Accurate End-to-End Span-based Semantic Role Labeling as Word-based Graph Parsing](https://mlsanity.com/paper/2112.02970/) (2021)
- [PriMeSRL-Eval: A Practical Quality Metric for Semantic Role Labeling Systems Evaluation](https://mlsanity.com/paper/2210.06408/) (2022)

## Abstract (arXiv)

Deep neural models achieve some of the best results for semantic role labeling. Inspired by instance-based learning that utilizes nearest neighbors to handle low-frequency context-specific training samples, we investigate the use of memory adaptation techniques in deep neural models. We propose a parameterized neighborhood memory adaptive (PNMA) method that uses a parameterized representation of the nearest neighbors of tokens in a memory of activations and makes predictions based on the most similar samples in the training data. We empirically show that PNMA consistently improves the SRL performance of the base model irrespective of types of word embeddings. Coupled with contextualized word embeddings derived from BERT, PNMA improves over existing models for both span and dependency semantic parsing datasets, especially on out-of-domain text, reaching F1 scores of 80.2, and 84.97 on CoNLL2005, and CoNLL2009 datasets, respectively.