What it does

This repository is a reading list: roughly 580-starred, manually curated collection of academic papers on event extraction in NLP. Papers are grouped by technique—pattern matching, machine learning, deep learning, semi-supervised and unsupervised learning, plus event coreference, surveys, datasets, and tools. Each entry includes a title, author, link, and often a pasted abstract.

The interesting bit

The chronological organization within categories makes the intellectual history visible. You can watch the field’s preoccupations shift from AutoSlog’s dictionary construction in 1993, through bootstrapping and multistrategy learning in the late 1990s, toward the modern era. It’s a bibliography with a narrative structure.

Key highlights

• Coverage from 1993 to present, with explicit technique-based categories
• Includes not just papers but also datasets, tools, and “other lists” for further rabbit-holing
• Abstracts are preserved in full, so you can skim without clicking through to PDFs
• Surveys section for quick orientation to subfields
• Event coreference gets its own section, often neglected in similar collections

Caveats

• No search, no tagging beyond the broad categories—finding a specific paper means scrolling or using browser find
• Star count (580) suggests utility but also that this is a niche reference; not a living community hub
• Abstracts appear to be copy-pasted directly from papers or Semantic Scholar, with occasional OCR artifacts (“perfomlance”)

Verdict

Worth bookmarking if you’re doing literature review in information extraction or teaching a graduate seminar. Skip it if you want code, benchmarks, or a systematic review with quality ratings—this is a well-organized pile of links, not a meta-analysis.