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

Retrieval-Augmented Question Answering on DeepSearch TriviaQA

78Success Rate (SR)

ActiveContext

Updated 5d ago

Evaluation Results

Method	Links
ActiveContext 2026.04		78	3.2
Full Context 2026.04		76	20.4
Training-Free GRPO 2026.04		76	16.7
Synapse 2026.04		68	51.2
ActiveContext 2026.04		66	3.7
ActiveContext 2026.04		66	3.6
Gemini-3.0-flash 2026.04		66	4.4
Training-Free GRPO 2026.04		64	18.3
Gemini-2.5-flash 2026.04		64	4.6
Mem0 2026.04		62	27.6
ContextCurator 2026.04		62	5
Training-Free GRPO 2026.04		60	38.4
Full Context 2026.04		58	20.6
Full Context 2026.04		58	20.7
Training-Free GRPO 2026.04		58	19.6
gpt-4o 2026.04		58	6.9
AWM 2026.04		56	16.4
gpt-4o-mini 2026.04		56	7.3
Qwen2.5-7B-Instruct 2026.04		52	6.5
Full Context 2026.04		50	19.7
Synapse 2026.04		37	42.7
Synapse 2026.04		36	42
Synapse 2026.04		35	86.1