List your company

Long Context Evaluation on Long Context Benchmarks

32.3MDQA-10 ScoreDD
24.08426.21728.3530.483May 21, 2024

Evaluation results

MethodLinks
DDStep Time (ms)=244, Data Prep. Cost=$, Curriculum=Grow-P2 with 8 cycles2024.0532.326.924.630.734.230.9
Baseline-8kStep Time (ms)=304, Data Prep. Cost=$, Document Masking=false2024.052923.820.526.23227.5
ICLMStep Time (ms)=304, Data Prep. Cost=$$$2024.0526.7202228.734.628.7
Pack-8k+DMStep Time (ms)=304, Data Prep. Cost=$$, Document Masking=true2024.0525.619.816.929.233.127.7
Baseline-8k+DMStep Time (ms)=304, Data Prep. Cost=$, Document Masking=true2024.0524.4201629.23227.1