# Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs

arXiv: https://arxiv.org/abs/2603.23532 · HTML: https://mlsanity.com/paper/2603.23532/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 12 of its 17 references that are in the map.
- First posted on arXiv 8 Mar 2026.

## Builds on (most cited first, 12 of 12)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Information Leakage in Embedding Models](https://mlsanity.com/paper/2004.00053/) (2020)
- [Structured information extraction from complex scientific text with fine-tuned large language models](https://mlsanity.com/paper/2212.05238/) (2022)
- [StructEval: Benchmarking LLMs' Capabilities to Generate Structural Outputs](https://mlsanity.com/paper/2505.20139/) (2025)
- [StructuredRAG: JSON Response Formatting with Large Language Models](https://mlsanity.com/paper/2408.11061/) (2024)
- [A Double-Graph Based Framework for Frame Semantic Parsing](https://mlsanity.com/paper/2206.09158/) (2022)
- [Rethinking the Privacy of Text Embeddings: A Reproducibility Study of "Text Embeddings Reveal (Almost) As Much As Text"](https://mlsanity.com/paper/2507.07700/) (2025)
- [Comparative Analysis of OpenAI GPT-4o and DeepSeek R1 for Scientific Text Categorization Using Prompt Engineering](https://mlsanity.com/paper/2503.02032/) (2025)
- [SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs](https://mlsanity.com/paper/2510.01248/) (2025)
- [Multi-Intent Recognition in Dialogue Understanding: A Comparison Between Smaller Open-Source LLMs](https://mlsanity.com/paper/2509.10010/) (2025)
- [Entity-Aware Self-Attention and Contextualized GCN for Enhanced Relation Extraction in Long Sentences](https://mlsanity.com/paper/2409.13755/) (2024)
- [HyperSumm-RL: A Dialogue Summarization Framework for Modeling Leadership Perception in Social Robots](https://mlsanity.com/paper/2507.04160/) (2025)

## Abstract (arXiv)

This paper investigates whether structured representations can preserve the meaning of scientific sentences. To test this, a lightweight LLM is fine-tuned using a novel structural loss function to generate hierarchical JSON structures from sentences collected from scientific articles. These JSONs are then used by a generative model to reconstruct the original text. Comparing the original and reconstructed sentences using semantic and lexical similarity we show that hierarchical formats are capable of retaining information of scientific texts effectively.