UniversalNER: Targeted Distillation from Large Language Models for Open Named Entity Recognition

7 Aug 2023  ·  Wenxuan Zhou, Sheng Zhang, Yu Gu, Muhao Chen, Hoifung Poon ·

Large language models (LLMs) have demonstrated remarkable generalizability, such as understanding arbitrary entities and relations. Instruction tuning has proven effective for distilling LLMs into more cost-efficient models such as Alpaca and Vicuna. Yet such student models still trail the original LLMs by large margins in downstream applications. In this paper, we explore targeted distillation with mission-focused instruction tuning to train student models that can excel in a broad application class such as open information extraction. Using named entity recognition (NER) for case study, we show how ChatGPT can be distilled into much smaller UniversalNER models for open NER. For evaluation, we assemble the largest NER benchmark to date, comprising 43 datasets across 9 diverse domains such as biomedicine, programming, social media, law, finance. Without using any direct supervision, UniversalNER attains remarkable NER accuracy across tens of thousands of entity types, outperforming general instruction-tuned models such as Alpaca and Vicuna by over 30 absolute F1 points in average. With a tiny fraction of parameters, UniversalNER not only acquires ChatGPT's capability in recognizing arbitrary entity types, but also outperforms its NER accuracy by 7-9 absolute F1 points in average. Remarkably, UniversalNER even outperforms by a large margin state-of-the-art multi-task instruction-tuned systems such as InstructUIE, which uses supervised NER examples. We also conduct thorough ablation studies to assess the impact of various components in our distillation approach. We will release the distillation recipe, data, and UniversalNER models to facilitate future research on targeted distillation.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Named Entity Recognition ACE 2005 UniNER-7B F1 86.69 # 1
Named Entity Recognition AnatEM UniNER-7B F1 88.65 # 1
Named Entity Recognition BC2GM UniNER-7B F1 82.42 # 1
Named Entity Recognition BC4CHEMD UniNER-7B F1 89.21 # 1
Named Entity Recognition BC5CDR UniNER-7B F1 89.34 # 1
Named Entity Recognition CoNLL03 UniNER-7B F1 93.3 # 1
Named Entity Recognition FindVehicle UniNER-7B F1 98.3 # 1
Named Entity Recognition GENIA UniNER-7B F1 77.54 # 1
Named Entity Recognition NCBI Disease UniNER-7B F1 86.96 # 1

Methods


No methods listed for this paper. Add relevant methods here