# Do LLMs Encode Frame Semantics? Evidence from Frame Identification

arXiv: https://arxiv.org/abs/2509.19540 · HTML: https://mlsanity.com/paper/2509.19540/

## Key facts

- Cited by 3 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 29 references that are in the map.
- First posted on arXiv 23 Sep 2025.

## Builds on (most cited first, 6 of 6)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Few-Shot Semantic Parsing with Language Models Trained On Code](https://mlsanity.com/paper/2112.08696/) (2021)
- [Frame-Semantic Parsing with Softmax-Margin Segmental RNNs and a Syntactic Scaffold](https://mlsanity.com/paper/1706.09528/) (2017)
- [Learning Joint Semantic Parsers from Disjoint Data](https://mlsanity.com/paper/1804.05990/) (2018)
- [Potential and Limitations of LLMs in Capturing Structured Semantics: A Case Study on SRL](https://mlsanity.com/paper/2405.06410/) (2024)
- [Coarse-to-Fine Dual Encoders are Better Frame Identification Learners](https://mlsanity.com/paper/2310.13316/) (2023)

## Built on by (most cited first, 3 of 3)

- [Evaluating the Impact of LLM-Assisted Annotation in a Perspectivized Setting: the Case of FrameNet Annotation](https://mlsanity.com/paper/2510.25904/) (2025)
- [How Prompts Move Language Model Behavior: Frames, Salience, and Construal as Semantic Control](https://mlsanity.com/paper/2512.12688/) (2025)
- [The Multilingual FrameNet Corpus](https://mlsanity.com/paper/2608.23037/) (2026)

## Abstract (arXiv)

We investigate whether large language models encode latent knowledge of frame semantics, focusing on frame identification, a core challenge in frame semantic parsing that involves selecting the appropriate semantic frame for a target word in context. Using the FrameNet lexical resource, we evaluate models under prompt-based inference and observe that they can perform frame identification effectively even without explicit supervision. To assess the impact of task-specific training, we fine-tune the model on FrameNet data, which substantially improves in-domain accuracy while generalizing well to out-of-domain benchmarks. Further analysis shows that the models can generate semantically coherent frame definitions, highlighting the model's internalized understanding of frame semantics.