# Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs

Source: [Generating Hierarchical JSON Representations of Scientific Sentences Using LLMs](<https://www.sota2.com/research/paper/69c4bd620636e5bf4afc7414>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-03-27T05:00:04.194Z

Authors: Satya Sri Rajiteswari Nimmagadda, Ethan Young, Niladri Sengupta, Ananya Jana, Aniruddha Maiti

Published: 2026-03-08T00:12:44.000Z

## Abstract

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.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Sentence Reconstruction on Scientific sentences n=274 (test)](<https://www.sota2.com/research/sota/sentence-reconstruction-on-scientific-sentences-n-274-test.md>): Sentence Reconstruction; Scientific sentences n=274 (test); Cosine Similarity: 0.8724.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.23532>)

[PDF](<https://arxiv.org/pdf/2603.23532.pdf>)
