New GRASP framework uses graph reasoning for academic literature reviews

Researchers have developed GRASP, a new framework designed to improve the generation of related work sections in academic papers. GRASP integrates Large Language Models with graph algorithms to map the complex relationships between cited sources. It utilizes a two-layer graph structure, including a Graph of Thoughts and an Argument-Counterargument Planning Network, to represent papers at varying levels of detail. The system employs Steiner tree algorithms for topology-aware pruning to identify crucial inter-paper connections, aiming to produce related work sections that closely mirror human-written examples in discourse roles and citation intent. AI

IMPACT This framework could streamline academic writing by automating the generation of literature reviews, allowing researchers to focus more on novel contributions.

RANK_REASON The item describes a new academic paper detailing a novel framework for academic research. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CL →

• alphaXiv
• Argument-Counterargument Planning Network
• arXivLabs
• CatalyzeX Code Finder for Papers
• Connected Papers
• CORE Recommender
• DagsHub
• Gotit.pub
• Graph of Thoughts
• Hugging Face
• Influence Flower
• Litmaps
• ScienceCast
• scite Smart Citations
• Steiner tree problem

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →