Share your thoughts, 1 month free Claude Pro on usSee more

Question Answering on SQuAD v2 (Corr., Err.)

1,853Correlation

LLM-RAG

Updated 14d ago

Evaluation Results

Method	Links
LLM-RAG 2026.05		1,853	11.93
LLM-RAG 2026.05		1,853	11.93
LLM-RAG 2026.05		1,850	11.75
BalanceRAG 2026.05		1,843	12.4
BalanceRAG 2026.05		1,842	12.41
UCB-Cascaded 2026.05		1,837	12.69
BalanceRAG 2026.05		1,836	12.1
UCB-Cascaded 2026.05		1,835	12.63
LLM-RAG 2026.05		1,831	10.95
BalanceRAG 2026.05		1,819	11.18
LLM-RAG 2026.05		1,803	10
BalanceRAG 2026.05		1,784	10.1
LLM-RAG 2026.05		1,755	14.16
BalanceRAG 2026.05		1,745	13.75
UCB-Cascaded 2026.05		1,744	13.54
LLM-RAG 2026.05		1,714	13.16
BalanceRAG 2026.05		1,699	12.71
LLM-RAG 2026.05		1,666	12.16
BalanceRAG 2026.05		1,654	11.64
LLM-RAG 2026.05		1,617	11.12
BalanceRAG 2026.05		1,591	10.66
LLM-RAG 2026.05		1,546	10.08
BalanceRAG 2026.05		1,484	9.75
LLM-only 2026.05		211	13.24
LLM-only 2026.05		194	13.19
LLM-only 2026.05		137	13.45