About PaperTrend

What PaperTrend does

PaperTrend tracks citation growth patterns of academic papers. For each paper we show its current growth state (e.g. accelerating, surging, plateau), a citation-over-time chart, and links to papers that cite it or that it references. All of this is computed from real citation data—we do not write manual summaries or generate interpretive text.

Data source and methodology

Citation data is collected from academic sources (e.g. Semantic Scholar and similar APIs). Paper metadata (title, authors, abstract, venue), citation counts over time, and reference/citation relationships are ingested and stored in our database. The numbers you see are not made up—they come from these external sources and are processed by our batch pipeline.

Growth state (e.g. Dormant, Slow growth, Accelerating, Surging, Plateau) is computed from citation metrics: we compare recent citation rate (e.g. last 30 days) to a prior baseline (e.g. prior 90 days) and assign a state label. Transitions (e.g. “from Steady growth to Surging”) are recorded when the state changes so you can see when a paper’s visibility shifted.

Update cycle

Data is updated by a separate batch process (not in real time when you load a page). New papers, citation metrics, and growth states are ingested and recomputed on a schedule (e.g. daily or weekly, depending on deployment). The web site you are viewing is read-only and caches pages for a short period (e.g. 30 minutes), so you may see data that is a few minutes to a few hours old.

Trust and transparency

We do not use AI-generated summaries or subjective interpretations. What you see is structured data: numbers, dates, state labels, and links to related papers. For privacy and data practices, see our Privacy page.