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

Multi-hop Question Answering on Bamboogle (test)

67.5EM

ARBOR

Updated 1mo ago

Evaluation Results

Method
ARBOR 2026.06	67.5	78.6	75.8
DAPO 2026.06	65.8	74.9	74.1
ARBOR 2026.06	65.3	76.5	72.9
ARBOR 2026.06	65.1	76.2	73
DAPO 2026.06	63.7	74.2	72.7
DAPO 2026.06	61.9	73.1	70.1
GRPO 2026.06	61.9	72.8	71.5
GRPO 2026.06	61.6	71	69.8
GRPO 2026.06	61	72.2	69.4
Workflow-R1-Search 2026.02	57.6	-	-
ExSearch 2026.04	51.27	56.12	63.8
AFlow 2026.02	50.4	-	-
CoT 2026.02	48	-	-
Workflow-R1 2026.02	47.2	-	-
HERA (Qwen-3-14B) 2026.04	46.46	49.01	60.53
TIR Prompting 2026.06	46.4	61.1	55.3
HERA (Llama-3.1-8) 2026.04	45.22	46.6	56.31
AceSearcher-8B 2026.04	45.12	47.21	51.91
SC (CoT×5) 2026.02	44.8	-	-
MedPrompt 2026.02	44.8	-	-
OThink-SRR1 2026.03	44	-	53.69
SE-Search-3B 2026.02	42.4	-	-
ReSearch 2026.03	42.4	-	53.65
TIR Prompting 2026.06	42.1	49.9	50.3
Direct 2026.02	40.8	-	-
Search-R1 (GRPO) 2026.02	40	-	-
Search-R1 2026.03	39.2	-	49.71
R1-Searcher 7B 2026.04	38.85	41	46.76
MaAS 2026.02	38.4	-	-
QCompiler 2025.05	37.1	41.9	49.2
Search-R1 (PPO) 2026.02	36.8	-	-
CriticSearch 2026.02	36.8	-	-
InForage 2026.02	36	-	-
GPT-4o-mini 2026.04	36	38.36	42.81
Qwen2.5-14B-Instruct 2026.04	34.7	27.2	34.35
O2-Searcher 2026.02	34.4	-	-
AutoRefine-Base 2026.02	34.4	-	-
DeepResearcher-7B 2026.04	34.2	37.42	44.4
Iter-RetGen 2026.04	34.12	35.33	40.65
AutoRefine-Instruct 2026.02	33.6	-	-
StepSearch-Base 2026.02	32.8	-	-
Search-r1-7B 2026.04	32.64	35.79	43.93
Search-o1 2026.02	32	-	-
MMAO-RAG-8B 2026.04	31.76	34.47	42.01
CORAG-8B (Greedy) 2026.04	31.61	33.03	41.25
OThink-SRR1 2026.03	31.2	-	43.58
TIR Prompting 2026.06	31.2	41.3	38.1
IRCoT 2026.04	30.73	34.88	33.97
Qwen3-14B 2026.04	29.6	31.54	38.31
Search-R1 2026.03	28	-	38.12
ReSearch-Instruct 2026.02	26.6	-	-
ReSearch 2026.03	26.4	-	36.45
RQ-RAG 2025.05	24.8	26.4	34.7
IRCoT 2026.02	24	-	-
Plan-RAG 2026.04	23.57	30.01	32.12
InstructRAG-8B 2026.04	23.44	26.16	31.18
Iter-Retgen 2025.05	23.2	24	31.8
Search-R1-Instruct 2026.02	23.2	-	-
SELF-RAG 2026.04	22.66	25.8	30.05
R1-Base 2026.02	22.4	-	-
Qwen3-8B 2026.04	22.4	23.87	30.28
IRCOT 2026.03	22.4	-	34.54
DAVID-GRPO 2026.01	22	-	25.4
IR-CoT 2025.05	21.6	26.4	34.2
IRCOT 2026.03	20.8	-	31.47
Basic RAG 2026.03	20	-	29.34
Iter-RetGen 2026.03	20	-	29.32
R1-Instruct 2026.02	19.2	-	-
Llama3-3.1-8B-instruct 2026.04	19.2	20.46	27.24
Naive RAG 2025.05	18.4	21.6	28.3
DAVID-GRPO 2026.01	14.4	-	24.2
ZeroSearch-Base 2026.02	13.9	-	-
Naive RAG 2025.05	13.6	15.2	22.1
Tree-GRPO 2026.01	13.6	-	-
Search-o1 2026.01	13	-	-
Search-R1-Base 2026.02	12.8	-	-
ReSearch-Base 2026.02	12.8	-	-
Tree-GRPO 2026.01	12	-	15.5
SFT 2026.02	11.2	-	-
Iter-RetGen 2026.03	11.2	-	19.09
No RAG 2026.03	10.4	-	17.57
Basic RAG 2026.03	9.6	-	19.06
Tree-GRPO 2026.01	9	-	11.9
Qwen2.5-7B-Instruct 2025.05	8.8	14.4	16.9
Direct Inference 2026.01	8	-	-
Search-R1-v0.3 2026.01	8	-	14.3
DAVID-GRPO 2026.01	8	-	14.5
Naive RAG 2026.02	8	-	-
DAVID-GRPO 2026.01	6	-	7
Tree-GRPO 2026.01	5	-	8.2
DAVID-GRPO 2026.01	4.8	-	8.1
GRPO 2026.01	4	-	-
Tree-GRPO 2026.01	4	-	8.9
StepSearch 2026.01	4	-	11
StepSearch 2026.01	4	-	6
Tree-GRPO 2026.01	4	-	5.5
Self-RAG 2025.05	3.2	16.8	13.1
StepSearch 2026.01	3.2	-	8.7
Tree-GRPO 2026.01	3.2	-	5.2
Search-R1-v0.3 2026.01	3	-	5.5

Showing 100 of 147 rows