Researchers have introduced SciTraj, a novel corpus designed to analyze the evolution of scientific research by providing a claim-grounded typed citation graph. Unlike traditional citation graphs that use a single edge type, SciTraj links citations to specific claim sentences within papers, categorizing them into six relation types. This corpus, covering 32,559 papers in NLP, ML, and Computer Vision from 2015-2024, reveals disciplinary siloing and highlights topic emergence, particularly in computer vision and large language model-related research. AI
IMPACT Provides a new tool for understanding research trends and interdisciplinary connections in AI fields.
RANK_REASON The cluster describes a new corpus and methodology for analyzing scientific research evolution, which falls under research. [lever_c_demoted from research: ic=1 ai=1.0]
Read on Hugging Face Daily Papers →
AI-generated summary · Google Gemini · from 1 sources. How we write summaries →