Extracting Polymer Nanocomposite Samples from Full-Length Documents
Not on the map. Why?
Abstract (arXiv)
This paper investigates the use of large language models (LLMs) for extracting sample lists of polymer nanocomposites (PNCs) from full-length materials science research papers. The challenge lies in the complex nature of PNC samples, which have numerous attributes scattered throughout the text. The complexity of annotating detailed information on PNCs limits the availability of data, making conventional document-level relation extraction techniques impractical due to the challenge in creating comprehensive named entity span annotations. To address this, we introduce a new benchmark and an evaluation technique for this task and explore different prompting strategies in a zero-shot manner. We also incorporate self-consistency to improve the performance. Our findings show that even advanced LLMs struggle to extract all of the samples from an article. Finally, we analyze the errors encountered in this process, categorizing them into three main challenges, and discuss potential strategies for future research to overcome them.
Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 18 of its 41 references that are in the map. First posted on arXiv 1 Mar 2024.
Builds on
18 of 41 references.
References · 18
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2023Llama 2: Open Foundation and Fine-Tuned Chat Modelscited by 12,865
- 2022Self-Consistency Improves Chain of Thought Reasoning in Language Modelscited by 5,601
- 2021Large Dual Encoders Are Generalizable Retrieverscited by 454
- 2023GPT-NER: Named Entity Recognition via Large Language Modelscited by 158
- 2023Does Synthetic Data Generation of LLMs Help Clinical Text Mining?cited by 151
- 2023Focused Transformer: Contrastive Training for Context Scalingcited by 133
- 2020SciREX: A Challenge Dataset for Document-Level Information Extractioncited by 111
- 2017Cross-Sentence N-ary Relation Extraction with Graph LSTMscited by 101
- 2023Revisiting Relation Extraction in the era of Large Language Modelscited by 100
- 2019Document-Level N-ary Relation Extraction with Multiscale Representation Learningcited by 52
- 2019Identification of Tasks, Datasets, Evaluation Metrics, and Numeric Scores for Scientific Leaderboards Constructioncited by 50
6 more
- 2023Aligning Instruction Tasks Unlocks Large Language Models as Zero-Shot Relation Extractorscited by 44
- 2023MatSci-NLP: Evaluating Scientific Language Models on Materials Science Language Tasks Using Text-to-Schema Modelingcited by 35
- 2023Large Language Models as Master Key: Unlocking the Secrets of Materials Science with GPTcited by 29
- 2022A sequence-to-sequence approach for document-level relation extractioncited by 25
- 2021CitationIE: Leveraging the Citation Graph for Scientific Information Extractioncited by 19
- 2023Reconstructing Materials Tetrahedron: Challenges in Materials Information Extractioncited by 8
Built on by
Citing papers per year: 2026: 1.
1 paper.
Similar, not cited
Close in wording, no citation either way
- 2024Mining experimental data from Materials Science literature with Large Language Models: an evaluation study
- 2026Retrieval Augmented Generation of Literature-derived Polymer Knowledge: The Example of a Biodegradable Polymer Expert System
- 2024MatViX: Multimodal Information Extraction from Visually Rich Articles
- 2025Fine-Tuning Vision-Language Models for Multimodal Polymer Property Prediction
- 2024Large language models, physics-based modeling, experimental measurements: the trinity of data-scarce learning of polymer properties
- 2024Are LLMs Ready for Real-World Materials Discovery?
- 2024AMGPT: a Large Language Model for Contextual Querying in Additive Manufacturing
- 2025Large language models in materials science and the need for open-source approaches