List your company

Question Answering (EM) on TriviaQA (test)

73.3EMWorkflow-R1-Search
24.10836.87949.6562.421Feb 1, 2026Feb 22, 2026Mar 16, 2026Apr 6, 2026Apr 28, 2026May 19, 2026Jun 10, 2026

Evaluation results

MethodLinks
Workflow-R1-SearchMulti-turn=true, Backbone=Qwen2.5-32B-Instruct2026.0273.3
ReCalRouting Strategy=RL-based Routing, Task Decomposition=false2026.0671.2
ReCal+Routing Strategy=RL-based Routing, Task Decomposition=false2026.0669.8
Router-R1Routing Strategy=RL-based Routing, Task Decomposition=false2026.0666.8
Search-R1 (GRPO)Multi-turn=true, Backbone=Qwen2.5-32B-Instruct2026.0262.3
Workflow-R1Multi-turn=true, Backbone=Qwen2.5-32B-Instruct2026.0261.2
Search-R1 (PPO)Multi-turn=true, Backbone=Qwen2.5-32B-Instruct2026.0261
MedPromptMulti-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0260.3
AFlowMulti-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0260.3
SC (CoT×5)Multi-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0259.5
RouterDCRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0659.2
GraphRouterRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0658.6
Prompt LLMRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0658
CoTMulti-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0257.9
Largest LLMRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0657.8
MaASMulti-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0257.1
RAGRouting Strategy=No Routing, Task Decomposition=false2026.0654
KNN RouterRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0652.8
RouteLLMRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0651.6
Search-R1Routing Strategy=No Routing, Task Decomposition=false2026.0651
Prompt LLMRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=true2026.0650
KNN RouterRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=true2026.0647.8
DirectMulti-turn=false, Backbone=Qwen2.5-32B-Instruct2026.0247.1
MLP RouterRouting Strategy=Heuristic & Discriminative Routing, Task Decomposition=false2026.0646
SFTRouting Strategy=No Routing, Task Decomposition=false2026.0640
CoTRouting Strategy=No Routing, Task Decomposition=false2026.0635.8
VanillaRouting Strategy=No Routing, Task Decomposition=false2026.0626