List your company

Intent Prediction on PhotoChat (test)

65.6F1 ScoreDRIBER ChatGPT 0613
39.91246.58153.2559.919Oct 16, 2021Feb 15, 2022Jun 18, 2022Oct 19, 2022Feb 19, 2023Jun 22, 2023Oct 23, 2023

Evaluation results

MethodLinks
DRIBER ChatGPT 0613Training Type=Zero-shot, LLM Backbone=ChatGPT 06132023.1065.666.764.7
DRIBER Vicuna-13BTraining Type=Zero-shot, LLM Backbone=Vicuna-13B2023.1065.564.367.1
DRIBER ChatGPT 1106Training Type=Zero-shot, LLM Backbone=ChatGPT 11062023.106463.165.1
PaCETotal parameter count=338 million2023.0563.863.368
PaCETraining Type=Fine-tuned2023.1063.863.368
DRIBER Falcon-Instruct-40BTraining Type=Zero-shot, LLM Backbone=Falcon-Instruct-40B2023.1062.862.463.3
T5-3B2021.1058.9——
T5-3BParameter count=3B2023.0558.954.164.6
T5-3BTraining Type=Fine-tuned2023.1058.954.164.6
T5-baseModel variant=base2023.0558.158.257.9
T5-baseTraining Type=Fine-tuned2023.1058.158.257.9
Divterjoint learning=true, G pre-train=true, F pre-train=true2021.1056.2——
DivterF pre-train=false2021.1055.9——
Divterjoint learning=false2021.1055.6——
BERT-base2021.1053.2——
BERT-baseModel variant=base2023.0553.256.150.6
BERT-baseTraining Type=Fine-tuned2023.1053.256.150.6
ViLT2023.0552.455.458.9
ViLTTraining Type=Fine-tuned2023.1052.455.458.9
ALBERT-baseModel variant=base2023.0552.244.862.7
ALBERT-baseTraining Type=Fine-tuned2023.1052.244.862.7
S2S-TF2021.1047.6——
DivterG pre-train=false2021.1047.3——
DivterG pre-train=false, F pre-train=false2021.1047.1——
DRIBER GPT-4 1106Training Type=Zero-shot, LLM Backbone=GPT-4 11062023.1045.656.568.1
DRIBER LLaMa2-Chat-70BTraining Type=Zero-shot, LLM Backbone=LLaMa2-Chat-70B2023.1040.95563.3