Abstract: Context-based determination of text similarity is a fundamental computation task that enables identification and attribution of emerging topics and narratives within and across information ...