List your company
ResearchFormerly WizWand

Predicate Grounding on ProDG Cooking

100F1 ScoreSymbolizer
67.7676.1384.592.87Apr 20, 2026
Updated 6mo ago

Evaluation results

MethodLinks
SymbolizerModel=Gem.3.1-Pro2026.04100
SymbolizerModel=Gem.3.1-FL2026.04100
SYMBOLIZERBackbone=Gemini 3.1 Pro2026.04100
SYMBOLIZERBackbone=Gemini 3.1 Flash Lite2026.04100
SYMBOLIZERBackbone=Gem.3.1-Pro2026.04100
SYMBOLIZERBackbone=Gem.3.1-FL2026.04100
SYMBOLIZERBackbone=Mistral Small2026.0499.6
SymbolizerModel=Mistral Small2026.0496.4
SYMBOLIZERBackbone=Mistral Small2026.0496.2
ViLaInBackbone=GPT-4, Retry Strategy=no retry2026.0496
ViLaInBackbone=GPT-4, Retry Strategy=two retries2026.0496
ViLaInBackbone=Gemini 3.1 Flash Lite, Retry Strategy=no retry2026.0495.7
ViLaInBackbone=Gemini 3.1 Flash Lite, Retry Strategy=two retries2026.0495.7
ViLaInBackbone=GPT-4, Retry Setting=two retries2026.0493
ViLaInBackbone=Gem. 3.1-FL, Retry Setting=no retry2026.0492.3
ViLaInBackbone=Gem. 3.1-FL, Retry Setting=two retries2026.0492.3
ViLaInModel=GPT-4, Retries=two retries2026.0491
ViLaInBackbone=GPT-4, Retry Setting=no retry2026.0488
ViLaInModel=Gem. 3.1-FL, Retries=no retry2026.0484.6
ViLaInModel=Gem. 3.1-FL, Retries=two retries2026.0484.6
ViLaInModel=GPT-4, Retries=no retry2026.0469