Few Shot Ner On Few Nerd Inter

Metrics

5 way 1~2 shot

5 way 5~10 shot

Results

Performance results of various models on this benchmark

			Paper Title
MSDP	76.86±0.22	84.78±0.69	A Multi-Task Semantic Decomposition Framework with Task-specific Pre-training for Few-Shot NER
JCELRNER	71.62±0.44	76.14±0.20	-
NuNER	67.37±0.31	73.50±0.09	NuNER: Entity Recognition Encoder Pre-training via LLM-Annotated Data
HEProto	66.40±0.18	72.53±0.11	HEProto: A Hierarchical Enhancing ProtoNet based on Multi-Task Learning for Few-shot Named Entity Recognition
TadNER	64.83±0.14	72.12±0.12	Type-Aware Decomposed Framework for Few-Shot Named Entity Recognition
DecomposedMetaNER	64.75±0.35	71.49±0.47	Decomposed Meta-Learning for Few-Shot Named Entity Recognition
DecomposedMetaSL	62.09±0.93	71.26±0.15	Decomposed Meta-Learning for Few-Shot Sequence Labeling
ESD	59.29±1.25	69.06±0.80	An Enhanced Span-based Decomposition Method for Few-Shot Sequence Labeling
BERT-SparseLT + CONTaiNER	57.14	66.17	Language Model Pre-Training with Sparse Latent Typing
CONTaiNER	55.95	61.83	CONTaiNER: Few-Shot Named Entity Recognition via Contrastive Learning
StructShot	51.88±0.69	57.32±0.63	Few-NERD: A Few-Shot Named Entity Recognition Dataset
NNShot	47.24±1.00	55.64±0.63	Few-NERD: A Few-Shot Named Entity Recognition Dataset
ProtoBERT	38.83±1.49	58.79±0.44	Few-NERD: A Few-Shot Named Entity Recognition Dataset

0 of 13 row(s) selected.

Few Shot Ner On Few Nerd Inter | SOTA | HyperAI