List your company
ResearchFormerly WizWand

Named Entity Recognition on ACE05

89.1F1 ScoreGOLLIE
-3.501620.539244.5868.6208Oct 3, 2018Jan 15, 2020Apr 28, 2021Aug 10, 2022Nov 22, 2023Mar 5, 2025Jun 17, 2026
Updated 4mo ago

Evaluation results

MethodLinks
GOLLIEevaluation=supervised2024.1189.1——
Hierarchical Multi-task ModelSetup=(I), Model=NER + EMD + RE2018.1187.5187.3787.65
Hierarchical Multi-task ModelSetup=(J), Model=NER + EMD + CR2018.1187.587.6787.34
KnowCoder-Xevaluation=supervised2024.1187.49——
Hierarchical Multi-task ModelSetup=(A), Model=Full Model2018.1187.3687.5287.21
BERTevaluation=supervised2024.1187.3——
Hierarchical Multi-task ModelSetup=(B), Model=NER2018.1187.1287.2487
Strubell et al.Model=Strubell et al. 20172018.1186.99——
Hierarchical Multi-task ModelSetup=(F), Model=NER + EMD2018.1186.9187.586.32
BERT-MRCTraining Setting=standard2021.0186.9——
UniversalNERevaluation=supervised2024.1186.69——
Hierarchical Multi-task ModelSetup=(K), Model=EMD + NER2018.1186.4385.6787.19
E2H-largeModel Category=Unified Models, Model Scale=large, Large-scale continued pre-training=false2023.0586.25——
Hierarchical Multi-task ModelSetup=(L), Model=EMD + NER + RE + CR2018.1186.2185.7886.66
KnowCoderevaluation=supervised2024.1186.1——
LasUIE*Model Category=Unified Models, Large-scale continued pre-training=true2023.0586——
UIE*Model Category=Unified Models, Large-scale continued pre-training=true2023.0585.78——
T5-largeModel Category=Unified Models, Model Scale=large, Large-scale continued pre-training=false2023.0585.76——
DeBiasModel Category=Specialized Models, Large-scale continued pre-training=false2023.0584.93——
TANLTraining Setting=standard2021.0184.9——
TANLModel Category=Unified Models, Large-scale continued pre-training=false2023.0584.9——
E2H-baseModel Category=Unified Models, Model Scale=base, Large-scale continued pre-training=false2023.0584.83——
BARTNERModel Category=Specialized Models, Large-scale continued pre-training=false2023.0584.74——
TANLTraining Setting=multi-dataset2021.0184.4——
T5-baseModel Category=Unified Models, Model Scale=base, Large-scale continued pre-training=false2023.0584.16——
B2NERevaluation=supervised2024.1183.04——
GLINERevaluation=supervised2024.1182.8——
IEPILEevaluation=supervised2024.1181.86——
YAYI-UIEevaluation=supervised2024.1181.78——
InstructUIEevaluation=supervised2024.1179.94——
Neural Transition-based Modelw/s=14452018.1073——
Ju et al.2018.1072.2——
Neural Transition-based Model (w/o dropout layer)w/s=14402018.1072——
Neural Transition-based Model (w/o char-level LSTM)w/s=15462018.1071.9——
Neural Transition-based Model (w/o pre-trained embeddings)w/s=14522018.1071.5——
Katiyar and Cardie2018.1070.5——
Muis and Luw/s=2632018.1063.1——
Lu and Rothw/s=4542018.1062.5——
GPT-5 miniZero-shot=true2026.0454——
BCLSummary Model=Llama-3.1-8B, Inference Model=Llama-3.1-8B2026.0653.1——
BCLModel=Llama-3.1-8b, Token Cost=5012026.0653.1——
BCLSummary Model=Qwen-2.5-7B, Inference Model=GPT-3.5-turbo2026.0653——
BCLSummary Model=Llama-3.1-8B, Inference Model=GPT-3.5-turbo2026.0652.5——
Prior Best ZSZero-shot=true2026.0451.2——
GPT-3.5-turboSummary Model=-, Inference Model=GPT-3.5-turbo2026.0651.03——
BCLSummary Model=Qwen-2.5-7B, Inference Model=Qwen-2.5-7B2026.0646.94——
BCLModel=Qwen-2.5-7b, Token Cost=5012026.0646.94——
DiZiNERZero-shot=true2026.0445——
GuideNERModel=Llama-3.1-8b, Token Cost=5062026.0644.97——
JPT-4Bzero-shot=true2026.0444.6——
BCLModel=Pixtral-12B, Token Cost=5012026.0642.87——
GuideNERModel=Qwen-2.5-7b, Token Cost=5062026.0641.57——
One-shotModel=Llama-3.1-8b, Token Cost=3852026.0640.98——
One-shotModel=Pixtral-12B, Token Cost=3852026.0639.15——
GuideNERModel=Pixtral-12B, Token Cost=5062026.0637.64——
UniNER-7Bzero-shot=true2026.0436.9——
One-shotModel=Qwen-2.5-7b, Token Cost=3852026.0635.73——
BCLModel=Qwen-2.5-3b, Token Cost=5012026.0635.46——
ChatIEModel=Pixtral-12B, Token Cost=9422026.0629.28——
GuideNERModel=Qwen-2.5-3b, Token Cost=5062026.0627.43——
GLiNER-Lzero-shot=true2026.0427.3——
One-shotModel=Qwen-2.5-3b, Token Cost=3852026.0627.08——
ChatIEModel=Llama-3.1-8b, Token Cost=9422026.0626.91——
DSEnsemble=true2026.0424.9——
CodeIEModel=Qwen-2.5-3b, Token Cost=11722026.0623.7——
MACEEnsemble=true2026.0423.5——
CodeIEModel=Qwen-2.5-7b, Token Cost=11722026.0622.76——
GLADEnsemble=true2026.0422.6——
MVEnsemble=true2026.0422.1——
CodeIEModel=Pixtral-12B, Token Cost=11722026.0618.57——
ChatIEModel=Qwen-2.5-3b, Token Cost=9422026.0615.41——
ChatIEModel=Qwen-2.5-7b, Token Cost=9422026.0611.83——
CodeIEModel=Llama-3.1-8b, Token Cost=11722026.060.06——