Using LLMs for the Extraction and Normalization of Product Attribute Values

Benchmark Model Rank Results
attribute-value-extraction-on-wdc-paveGPT-4_10_example_values_&_10_demonstrations#1F1-Score: 90.54
attribute-value-extraction-on-wdc-paveGPT-3.5_10_example_values_&_10_demonstrations#2F1-Score: 88.02
attribute-value-extraction-on-wdc-paveAVEQA#3F1-Score: 80.83
attribute-value-extraction-on-wdc-paveMAVEQA#4F1-Score: 65.10
attribute-value-extraction-on-wdc-paveSU-OpenTag#5F1-Score: 60.44