List your company

Multi-hop Question Answering on MuSiQue 2022

51.2F1 ScoreLongLLMLingua
18.33626.86835.443.932Oct 10, 2023

Evaluation results

MethodLinks
LongLLMLinguaTarget LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1051.21,0772.3
LongLLMLingua (w/o Subsequence Recovery)Target LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1050.71,0772.3
LongLLMLingua (w/o restrict)Target LLM=GPT-3.5-Turbo, Compression constraint=2x, Module=Question-aware Coarse-grained2023.1049.21,0782.3
LongLLMLingua (w/o Document Reordering)Target LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1049.21,0772.3
LongLLMLingua (w/o Question-aware Fine-grained)Target LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1048.41,1182.2
LongLLMLingua (w/ GPT2-small)Target LLM=GPT-3.5-Turbo, Compression constraint=2x, Model=GPT2-small2023.1048.41,0952.2
LongLLMLingua (w/o Dynamic Compression Ratio)Target LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1048.21,0902.2
LongLLMLingua (w/ SBERT)Target LLM=GPT-3.5-Turbo, Compression constraint=2x, Module=Question-aware Coarse-grained2023.1047.31,0702.3
LongLLMLingua r_kTarget LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1046.31,2951.9
Original PromptTarget LLM=GPT-3.5-Turbo2023.1045.82,427—
LongLLMLingua (w/ p(x_doc|x_que, restrict))Target LLM=GPT-3.5-Turbo, Compression constraint=2x, Module=Question-aware Coarse-grained2023.10441,0662.3
LongLLMLingua (w/o Question-awareness)Target LLM=GPT-3.5-Turbo, Compression constraint=2x, Module=Question-aware Coarse-grained2023.1043.21,0762.3
LLMLinguaTarget LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1040.11,1102.2
SBERTTarget LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1036.21,2881.9
BM25Target LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1028.51,2951.9
Selective-ContextTarget LLM=GPT-3.5-Turbo, Compression constraint=2x2023.1019.61,1412.1