HomeAI & Machine LearningDependency Parsing 2D: Arc-Standard Shift-Reduce Parser

Dependency Parsing 2D: Arc-Standard Shift-Reduce Parser

Step or auto-run a transition-based dependency parser across a flat 2D stage: watch SHIFT, LEFT-ARC and RIGHT-ARC move words between a live stack panel and buffer queue while arcs animate above the sentence.

AI & Machine Learning2DModerate60 FPS📱 Mobile-adapted⇄ 3D version
2d-ai-topic-40 ↗ Open standalone

This is the flat, panel-based twin of the 3D dependency-parser simulator: the same arc-standard transition system — a stack of partially-processed words, a buffer of words still waiting to be read, and the three moves SHIFT, LEFT-ARC, RIGHT-ARC — rendered instead as three synchronized 2D canvases. The main stage draws the sentence with dependency arcs curving into place above it (pan by dragging, scale with the zoom slider); a dedicated stack panel and buffer panel mirror the parser's live internal state below, and a transition log records every move as it happens. Step through the oracle one transition at a time or auto-run it across five example sentences, and confirm for yourself that every projective tree here takes exactly 2n − 1 transitions to build.

⚙ Under the hood

Step or auto-run a transition-based dependency parser across three synchronized flat panels: the main stage draws the sentence with dependency arcs curving into place (drag to pan, zoom slider to scale), while dedicated stack and buffer panels mirror the parser's live state as SHIFT, LEFT-ARC and RIGHT-ARC transitions fire.

nlpdependency-parsingsyntaxtransition-basedlinguisticsalgorithms

2D · HTML5 Canvas 2D · 60 FPS target · runs fully client-side, no install

What did you find?

Add reproduction steps (optional)