# jiesutd/LatticeLSTM

Chinese NER using Lattice LSTM. Code for ACL 2018 paper.

Repository: https://github.com/jiesutd/LatticeLSTM
Canonical: https://ross.abutalabs.com/products/latticelstm
Language: Python
License Family: other
Topics: lattice-lstm, lstm-crf, ner, chinese-ner, lattice-lstm-crf
Last push: 2019-04-25T14:44:32+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 3055, "days_push": 2687, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1832, forks 446 (observed 2026-08-28T04:05:42.293035+00:00)

## What it is
A PyTorch implementation of Lattice LSTM for Chinese Named Entity Recognition, accompanying an ACL 2018 paper. It combines character-based LSTM with lattice word embeddings and achieved state-of-the-art F1 on the MSRA dataset at publication time.

## Use cases
- run Chinese named entity recognition on text
- train a Chinese NER model with lattice LSTM
- reproduce ACL 2018 Chinese NER paper results
- extract person, location, organization entities from Chinese text
- train NER on CoNLL-format BIOES data
- use pretrained character and word embeddings for Chinese NER

## When to choose
- you need to reproduce or extend the ACL 2018 Lattice LSTM paper
- you work with legacy Python 2.7 / PyTorch 0.3 environments
- you want a research baseline for character-based Chinese NER with word-lattice features

## When to avoid
- you need a maintained, production-ready NER toolkit
- you use modern PyTorch or Python 3
- you want transformer-based or multilingual NER models

## Facets
- artifact type: library
- maturity: abandoned
- function: nlp, machine-learning, deep-learning
- domain: machine-learning, deep-learning
- platform: python
- tags: named-entity-recognition, lstm-crf, lattice-lstm, chinese-ner, pytorch, research-code, acl-2018, natural-language-processing

## Member repositories
- jiesutd/LatticeLSTM (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:42.293035+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:19:09.667640+00:00, confidence not recorded.
  - readme: https://github.com/jiesutd/LatticeLSTM (fetched 2026-08-28T04:05:42.293035+00:00, sha 2a4f9d0dfb08)
- Data as of 2026-08-30T08:39:29.467469+00:00.