List your company
ResearchFormerly WizWand

Online Shopping on WebShop (test)

95ScoreSkillMaster
-3.821.8547.573.15May 9, 2026May 12, 2026May 15, 2026May 18, 2026May 21, 2026May 24, 2026May 28, 2026
Updated 5mo ago

Evaluation results

MethodLinks
SkillMasterCategory=Memory-Augmented RL-based Methods2026.059582
SERLType=Hybrid2026.0589.580.1
GraphGPOType=RL Training, Base Model=Qwen2.5-1.5B-Instruct2026.0589.2978.65
HGPOType=RL Training2026.0588.477.8
GRPO+SDPO(Loss)Type=Hybrid2026.0588.473
GiGPOType=RL Training, Base Model=Qwen2.5-1.5B-Instruct2026.0587.9473.83
GraphGPOType=RL Training, Base Model=Qwen2.5-7B-Instruct2026.0586.9480.31
GiGPOType=RL Training, Base Model=Qwen2.5-7B-Instruct2026.0586.7278.38
SkillRLCategory=Memory-Augmented RL-based Methods2026.0585.272.7
GRPO+SDPO(Advantage)Type=Hybrid2026.0584.875.4
GRPOType=RL Training, Base Model=Qwen2.5-1.5B-Instruct2026.0584.7371.35
GRPOType=RL Training, Base Model=Qwen2.5-7B-Instruct2026.0584.3175
RLSDType=Hybrid2026.0583.675.8
GIGPOType=RL Training2026.0583.575.8
PPO (with critic)Type=RL Training2026.0581.468.7
PPOType=RL Training, Base Model=Qwen2.5-7B-Instruct2026.0581.468.7
RLOOCategory=RL-based Methods2026.0580.365.7
RLOOType=RL Training2026.0580.365.7
RLOOType=RL Training, Base Model=Qwen2.5-7B-Instruct2026.0580.365.7
GRPOCategory=RL-based Methods2026.0579.366.1
RLOOType=RL Training, Base Model=Qwen2.5-1.5B-Instruct2026.0573.952.1
PPOType=RL Training, Base Model=Qwen2.5-1.5B-Instruct2026.0573.851.5
GRPOType=RL Training2026.0573.164.1
SimpleMem+GRPOCategory=Memory-Augmented RL-based Methods2026.0567.846.9
SkillBrewBackbone=Qwen2.5-7B-Instruct2026.0559.338.4
ReflexionCategory=Prompt-based Agentic or Memory-based Methods2026.0558.128.8
Mem0+GRPOCategory=Memory-Augmented RL-based Methods2026.0558.137.5
ReflexionType=Prompting2026.0558.128.8
ReflexionType=Prompting, Base Model=Qwen2.5-7B-Instruct2026.0558.128.8
ReflexionBackbone=Qwen2.5-7B-Instruct2026.0558.128.8
ReflexionType=Prompting, Base Model=Qwen2.5-1.5B-Instruct2026.0555.821.9
VoyagerBackbone=Qwen2.5-7B-Instruct2026.055026.4
ReActCategory=Prompt-based Agentic or Memory-based Methods2026.0546.219.5
ReActType=Prompting2026.0546.219.5
ReActType=Prompting, Base Model=Qwen2.5-7B-Instruct2026.0546.219.5
ReActBackbone=Qwen2.5-7B-Instruct2026.0546.219.5
Gemini-2.5-ProCategory=Closed-source LLMs2026.0542.535.9
EvolveRCategory=Memory-Augmented RL-based Methods2026.0542.517.6
Gemini-2.5-ProType=Prompting, Base Model=Closed-Source2026.0542.535.9
EvoSkillBackbone=Qwen2.5-7B-Instruct2026.0542.212.6
ReActType=Prompting, Base Model=Qwen2.5-1.5B-Instruct2026.0540.111.3
Skill-ProBackbone=Qwen2.5-7B-Instruct2026.0538.723.2
SimpleMemCategory=Prompt-based Agentic or Memory-based Methods2026.0533.28.59
SimpleMemBackbone=Qwen2.5-7B-Instruct2026.0533.28.59
GPT-4oCategory=Closed-source LLMs2026.0531.823.7
GPT-4oType=Prompting, Base Model=Closed-Source2026.0531.823.7
ExpeLCategory=Prompt-based Agentic or Memory-based Methods2026.0530.911.2
ExpeLBackbone=Qwen2.5-7B-Instruct2026.0530.911.2
MemRLCategory=Memory-Augmented RL-based Methods2026.0529.59.2
Qwen2.5-7B-InstructCategory=Prompt-based Agentic or Memory-based Methods2026.0526.47.8
Qwen2.5-7B-InstructType=Prompting2026.0526.47.8
Qwen2.5Type=Prompting, Base Model=Qwen2.5-7B-Instruct2026.0526.47.8
Zero-ShotBackbone=Qwen2.5-7B-Instruct2026.0526.47.8
MemPCategory=Prompt-based Agentic or Memory-based Methods2026.0525.36.4
MemPBackbone=Qwen2.5-7B-Instruct2026.0525.36.4
Mem0Category=Prompt-based Agentic or Memory-based Methods2026.0523.92
Mem0Backbone=Qwen2.5-7B-Instruct2026.0523.92
Qwen2.5Type=Prompting, Base Model=Qwen2.5-1.5B-Instruct2026.0523.15.2
SDPOType=Self-Distillation2026.0500