Retrieval-Augmented Question Answering

Benchmarks

Dataset Name	SOTA Method	Metric
NQ		Clean Accuracy89	45	2mo ago
RAGQA Leaderboard (test)	SFT+RL(Rhybrid)	AVG Score85	29	4mo ago
DeepSearch Average	ActiveContext	SR57.1	23	3mo ago
DeepSearch Bamboogle	Synapse	Success Rate (SR)72	23	3mo ago
DeepSearch Musique	Synapse	SR46	23	3mo ago
DeepSearch 2wiki	Training-Free GRPO	Success Rate (SR)68	23	3mo ago
DeepSearch HotpotQA	ActiveContext	Success Rate56	23	3mo ago
DeepSearch PopQA	ActiveContext	Success Rate64	23	3mo ago
DeepSearch TriviaQA	ActiveContext	Success Rate (SR)78	23	3mo ago
DeepSearch NQ	Synapse	SR86	23	3mo ago
NQ, TriviaQA, and WebQ (test)	ERA	Answer Quality Recall62.2	18	3mo ago
ClashEval	Gemma (v3.0, 1B)	Epsilon Obey70	16	2mo ago
NQ	PoisonedRAG	ATR100	16	3mo ago
MS-MARCO	PoisonedRAG	ATR98	16	3mo ago
HotpotQA	PoisonedRAG	ATR100	16	3mo ago
Unique Canonical User-Facing Queries n=170 (Overall)		Faithfulness97	12	1mo ago
Composite-9	Custom ReAct	Correctness94.4	10	1mo ago
170 unique canonical user-facing queries Complex Answer n=46	Naive RAG	Faithfulness88	6	1mo ago
170 unique canonical user-facing queries Complex Rec. n=30	I-GUIDE SS	Faithfulness72	6	1mo ago
170 unique canonical user-facing queries Simple Rec. n=50		Faithfulness96	6	1mo ago
MultiHop-RAG	Monolithic	Correctness (%)69.2	6	1mo ago
PopQA sampled queries from Augmented dataset	HaS	Average Latency (s)0.8725	6	3mo ago
Granola-EQ sampled queries from Augmented dataset	HaS	Average Latency (s)1.0559	6	3mo ago
MMLU-Pro	HYBRID-ROUTED	Correlation (%)51.8	4	1mo ago
GraphRAG-Bench 5-novel 519 questions	RLM-on-KG	Precision41.9	3	3mo ago

Showing 25 of 25 rows