List your company

Question Answering on HotpotQA (test) (Tool Use Metrics)

28.4EMUALA-M+Backoff
16.33619.46822.625.732Jan 25, 2024

Evaluation results

MethodLinks
UALA-M+BackoffBackbone=LLaMA2-13B2024.0128.41,525
UALA-S+BackoffBackbone=LLaMA2-13B2024.0127.3539
UALA-MBackbone=LLaMA2-13B2024.0126.51,525
SCBackbone=LLaMA2-13B2024.0126.40
UALA-SBackbone=LLaMA2-13B2024.0125.6539
ReAct+BackoffBackbone=LLaMA2-13B2024.01252,371
CoTBackbone=LLaMA2-13B2024.0124.80
UALA-M+BackoffBackbone=LLaMA2-7B2024.0124.61,386
UALA-S+BackoffBackbone=LLaMA2-7B2024.0124.2479
ReAct+BackoffBackbone=LLaMA2-7B2024.0123.82,324
StandardBackbone=LLaMA2-13B2024.0123.40
UALA-SBackbone=LLaMA2-7B2024.0122.8479
UALA-MBackbone=LLaMA2-7B2024.0122.61,386
SCBackbone=LLaMA2-7B2024.01220
CoTBackbone=LLaMA2-7B2024.0121.80
StandardBackbone=LLaMA2-7B2024.0121.20
ReActBackbone=LLaMA2-13B2024.0119.82,371
ReActBackbone=LLaMA2-7B2024.0116.82,324