Decomposed Meta-Learning for Few-Shot Named Entity Recognition

Few-shot named entity recognition (NER) systems aim at recognizing novel-class named entities based on only a few labeled examples. In this paper, we present a decomposed meta-learning approach which addresses the problem of few-shot NER by sequentially tackling few-shot span detection and few-shot entity typing using meta-learning. In particular, we take the few-shot span detection as a sequence labeling problem and train the span detector by introducing the model-agnostic meta-learning (MAML) algorithm to find a good model parameter initialization that could fast adapt to new entity classes. For few-shot entity typing, we propose MAML-ProtoNet, i.e., MAML-enhanced prototypical networks to find a good embedding space that can better distinguish text span representations from different entity classes. Extensive experiments on various benchmarks show that our approach achieves superior performance over prior methods.

PDF Abstract Findings (ACL) 2022 PDF Findings (ACL) 2022 Abstract

Results from the Paper

Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Few-shot NER Few-NERD (INTER) DecomposedMetaNER 5 way 1~2 shot 64.75±0.35 # 1
5 way 5~10 shot 71.49±0.47 # 1
10 way 1~2 shot 58.65±0.43 # 1
10 way 5~10 shot 68.11±0.05 # 1
Few-shot NER Few-NERD (INTRA) DecomposedMetaNER 5 way 1~2 shot 49.48±0.85 # 1
5 way 5~10 shot 62.92±0.57 # 1
10 way 1~2 shot 42.84±0.46 # 1
10 way 5~10 shot 57.31±0.25 # 1


No methods listed for this paper. Add relevant methods here