List your company

Web Navigation on WebShop (val)

85.3Success RateTSR (Beam Search)
-0.81221.54443.966.256Feb 12, 2026Feb 23, 2026Mar 6, 2026Mar 18, 2026Mar 29, 2026Apr 9, 2026Apr 21, 2026

Evaluation results

MethodLinks
TSR (Beam Search)Model=Qwen2.5-3B2026.0285.34535.8
TSR (Lookahead)Model=Qwen2.5-3B2026.0282.34756.1
TSR (Best-of-N)Model=Qwen2.5-3B2026.0277.35046.5
Instance FilteringModel=Qwen2.5-3B2026.0270.35196.8
EVPOBackbone=Qwen2.5-3B-Instruct2026.0430.3——
PPOBackbone=Qwen2.5-3B-Instruct, Method Category=Critic-based2026.0425.2——
StarPO-SBackbone=Qwen2.5-3B-Instruct, Method Category=Critic-based2026.0417.8——
GRPOBackbone=Qwen2.5-3B-Instruct, Method Category=Critic-free2026.0414.3——
DAPOBackbone=Qwen2.5-3B-Instruct, Method Category=Critic-free2026.0412.7——
Base ModelModel=Qwen2.5-3B2026.0237477.7
Base LLMBackbone=Qwen2.5-3B-Instruct, Method Category=None2026.042.5——