List your company
ResearchFormerly WizWand

Multi-hop Question Answering on Complex-TR ODQA 1.0 (test)

0.312Set AccuracyFiD-PIT + Refine
0.08840.146450.20450.26255Nov 16, 2023
Updated 8mo ago

Evaluation results

MethodLinks
FiD-PIT + RefineModel Backbone=T5-large, Architecture=FiD, Reasoning Enhancement=PIT, Context Refinement=true2023.110.3120.391
FiD-PITModel Backbone=T5-large, Architecture=FiD, Reasoning Enhancement=PIT2023.110.290.375
FiDModel Backbone=T5-large, Architecture=FiD2023.110.2740.373
PIT-SFTModel Backbone=T5-large, Architecture=SFT, Reasoning Enhancement=PIT2023.110.2530.328
FiD-PIT + RefineModel Backbone=T5-base, Architecture=FiD, Reasoning Enhancement=PIT, Context Refinement=true2023.110.2470.312
FiD-PITModel Backbone=T5-base, Architecture=FiD, Reasoning Enhancement=PIT2023.110.2310.302
SFTModel Backbone=T5-large, Architecture=SFT2023.110.2310.321
PIT-SFTModel Backbone=T5-base, Architecture=SFT, Reasoning Enhancement=PIT2023.110.2260.301
SFTModel Backbone=T5-base, Architecture=SFT2023.110.1770.266
FiDModel Backbone=T5-base, Architecture=FiD2023.110.1770.26
FLAN-T5-XLModel Backbone=FLAN-T5-XL2023.110.1450.203
GPT-4Model Backbone=GPT-42023.110.140.372
GPT-3.5Model Backbone=GPT-3.52023.110.0970.23