# Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRL

arXiv: https://arxiv.org/abs/2405.06410 · HTML: https://mlsanity.com/paper/2405.06410/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- Builds on 24 of its 36 references that are in the map.
- First posted on arXiv 10 May 2024.

## Builds on (most cited first, 24 of 24)

- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Training language models to follow instructions with human feedback](https://mlsanity.com/paper/2203.02155/) (2022)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [Chain-of-Thought Prompting Elicits Reasoning in Large Language Models](https://mlsanity.com/paper/2201.11903/) (2022)
- [The Power of Scale for Parameter-Efficient Prompt Tuning](https://mlsanity.com/paper/2104.08691/) (2021)
- [Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)](https://mlsanity.com/paper/1711.11279/) (2017)
- [What Does BERT Look At? An Analysis of BERT's Attention](https://mlsanity.com/paper/1906.04341/) (2019)
- [How Can We Know What Language Models Know?](https://mlsanity.com/paper/1911.12543/) (2019)
- [Exploiting Cloze Questions for Few Shot Text Classification and Natural Language Inference](https://mlsanity.com/paper/2001.07676/) (2020)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [What do you learn from context? Probing for sentence structure in contextualized word representations](https://mlsanity.com/paper/1905.06316/) (2019)
- [Assessing the Ability of LSTMs to Learn Syntax-Sensitive Dependencies](https://mlsanity.com/paper/1611.01368/) (2016)
- [Constrained Language Models Yield Few-Shot Semantic Parsers](https://mlsanity.com/paper/2104.08768/) (2021)
- [On the State of the Art of Evaluation in Neural Language Models](https://mlsanity.com/paper/1707.05589/) (2017)
- [Compositional Semantic Parsing with Large Language Models](https://mlsanity.com/paper/2209.15003/) (2022)
- [Few-Shot Text Generation with Pattern-Exploiting Training](https://mlsanity.com/paper/2012.11926/) (2020)
- [Few-Shot Semantic Parsing with Language Models Trained On Code](https://mlsanity.com/paper/2112.08696/) (2021)
- [Large-Scale QA-SRL Parsing](https://mlsanity.com/paper/1805.05377/) (2018)
- [Analyzing and Interpreting Neural Networks for NLP: A Report on the First BlackboxNLP Workshop](https://mlsanity.com/paper/1904.04063/) (2019)
- [Establishing Strong Baselines for the New Decade: Sequence Tagging, Syntactic and Semantic Parsing with BERT](https://mlsanity.com/paper/1908.04943/) (2019)
- [Parsing as Pretraining](https://mlsanity.com/paper/2002.01685/) (2020)
- [Semantic Role Labeling as Dependency Parsing: Exploring Latent Tree Structures Inside Arguments](https://mlsanity.com/paper/2110.06865/) (2021)
- [Unsupervised Induction of Semantic Roles within a Reconstruction-Error Minimization Framework](https://mlsanity.com/paper/1412.2812/) (2014)
- [An MRC Framework for Semantic Role Labeling](https://mlsanity.com/paper/2109.06660/) (2021)

## Built on by (most cited first, 10 of 10)

- [Semantic Mastery: Enhancing LLMs with Advanced Natural Language Understanding](https://mlsanity.com/paper/2504.00409/) (2025)
- [AURA: An Agent Autonomy Risk Assessment Framework](https://mlsanity.com/paper/2510.15739/) (2025)
- [Can LLMs Extract Frame-Semantic Arguments?](https://mlsanity.com/paper/2502.12516/) (2025)
- [Do LLMs Encode Frame Semantics? Evidence from Frame Identification](https://mlsanity.com/paper/2509.19540/) (2025)
- [Hierarchical Frequency Tagging Probe (HFTP): A Unified Approach to Investigate Syntactic Structure Representations in Large Language Models and the Human Brain](https://mlsanity.com/paper/2510.13255/) (2025)
- [Emergence and Localisation of Semantic Role Circuits in LLMs](https://mlsanity.com/paper/2511.20910/) (2025)
- [Multi-objective Representation for Numbers in Clinical Narratives: A CamemBERT-Bio-Based Alternative to Large-Scale LLMs](https://mlsanity.com/paper/2405.18448/) (2024)
- [LLMs Can Also Do Well! Breaking Barriers in Semantic Role Labeling via Large Language Models](https://mlsanity.com/paper/2506.05385/) (2025)
- [Beyond Understanding: Evaluating the Pragmatic Gap in LLMs' Cultural Processing of Figurative Language](https://mlsanity.com/paper/2510.23828/) (2025)
- [Lost in Reordering: Structural Sensitivity of Multilingual LLMs under Semantics-Preserving Perturbations](https://mlsanity.com/paper/2609.03511/) (2026)

## Abstract (arXiv)

Large Language Models (LLMs) play a crucial role in capturing structured semantics to enhance language understanding, improve interpretability, and reduce bias. Nevertheless, an ongoing controversy exists over the extent to which LLMs can grasp structured semantics. To assess this, we propose using Semantic Role Labeling (SRL) as a fundamental task to explore LLMs' ability to extract structured semantics. In our assessment, we employ the prompting approach, which leads to the creation of our few-shot SRL parser, called PromptSRL. PromptSRL enables LLMs to map natural languages to explicit semantic structures, which provides an interpretable window into the properties of LLMs. We find interesting potential: LLMs can indeed capture semantic structures, and scaling-up doesn't always mirror potential. Additionally, limitations of LLMs are observed in C-arguments, etc. Lastly, we are surprised to discover that significant overlap in the errors is made by both LLMs and untrained humans, accounting for almost 30% of all errors.