List your company

One-step next-observation prediction on AlfWorld (test)

89Token F1Word2World
33.8848.1962.576.81May 29, 2026

Evaluation results

MethodLinks
Word2WorldBackbone=Qwen3.5-4B, LLM Usage=SFT, Infer. LLM?=Yes2026.058966
PatchWorld-ResidualBackbone=Mimo-v2.5, LLM Usage=CodeGen, Infer. LLM?=No2026.057750
PatchWorld-SimpleBackbone=Mimo-v2.5, LLM Usage=CodeGen, Infer. LLM?=No2026.057347
PatchWorld-ResidualBackbone=Qwen3-Coder-480B, LLM Usage=CodeGen, Infer. LLM?=No2026.057047
WorldCoderBackbone=Qwen3-Coder-480B, LLM Usage=CodeGen, Infer. LLM?=No2026.056342
PoE-WorldBackbone=Qwen3-Coder-480B, LLM Usage=CodeGen, Infer. LLM?=No2026.056240
PoE-WorldBackbone=Mimo-v2.5, LLM Usage=CodeGen, Infer. LLM?=No2026.056240
WorldCoderBackbone=Mimo-v2.5, LLM Usage=CodeGen, Infer. LLM?=No2026.055935
PatchWorld-ResidualBackbone=DeepSeek-V4-Flash, LLM Usage=CodeGen, Infer. LLM?=No2026.055729
LLM-DirectBackbone=Mimo-v2.5, LLM Usage=ICL, Infer. LLM?=Yes2026.055533
LLM-DirectBackbone=DeepSeek-V4-Flash, LLM Usage=ICL, Infer. LLM?=Yes2026.055534
LLM-DirectBackbone=Qwen3-Coder-480B, LLM Usage=ICL, Infer. LLM?=Yes2026.055335
PatchWorld-SimpleBackbone=DeepSeek-V4-Flash, LLM Usage=CodeGen, Infer. LLM?=No2026.054821
WorldCoderBackbone=DeepSeek-V4-Flash, LLM Usage=CodeGen, Infer. LLM?=No2026.054021
PoE-WorldBackbone=DeepSeek-V4-Flash, LLM Usage=CodeGen, Infer. LLM?=No2026.053714
PatchWorld-SimpleBackbone=Qwen3-Coder-480B, LLM Usage=CodeGen, Infer. LLM?=No2026.053610